Preface.........................................................................................................................xxix
I General Information.....................................................................................................1
1 Introduction..................................................................................................................3
1.1 What is this book about?......................................................................................3
1.2 Which models are considered?.............................................................................4
1.3 Whom is this book for?.........................................................................................5
1.4 How is the book organized?.................................................................................5
1.5 What software do you need?.................................................................................7
1.5.1 Updating Stata 9............................................................................................8
1.5.2 Installing SPost..............................................................................................9
Installing SPost using search.......................................................................9
Installing SPost using net install................................................................11
1.5.3 What if commands do not work?................................................................11
1.5.4 Uninstalling SPost.........................................................................................12
1.5.5 Using spex to load data and run examples...............................................12
1.5.6 More files available on the web site..........................................................13
1.6 Where can I learn more about the models?.........................................................13
2 Introduction to Stata....................................................................................................15
2.1 The Stata interface.................................................................................................16
Changing the scrollback buffer size.......................................................18
Changing the display of variable names in the Variables window ....19
2.2 Abbreviations...........................................................................................................19
2.3 How to get help......................................................................................................20
2.3.1 Online help.....................................................................................................20
2.3.2 Manuals..........................................................................................................21
2.3.3 Other resources.............................................................................................21
2.4 The working directory...........................................................................................21
2.5 Stata file types........................................................................................................22
2.6 Saving output to log files......................................................................................22
Options......................................................................................................23
2.6.1 Closing a log file..........................................................................................23
2.6.2 Viewing a log file........................................................................................24
2.6.3 Converting from SMCL to plain text or PostScript ................................24
2.7 Using and saving datasets....................................................................................24
2.7.1 Data in Stata format....................................................................................24
2.7.2 Data in other formats..................................................................................25
2.7.3 Entering data by hand.................................................................................26
2.8 Size limitations on datasets*................................................................................26
2.9 Do-files...................................................................................................................26
2.9.1 Adding comments.........................................................................................28
2.9.2 Long lines.....................................................................................................29
2.9.3 Stopping a do-file while it is running.......................................................29
2.9.4 Creating do-files..........................................................................................29
Using Stata's Do-file Editor.......................................................................29
Using other editors to create do-files......................................................30
2.9.5 Recommended structure for do-files.........................................................30
2.10 Using Stata for serious data analysis................................................................31
2.11 Syntax of Stata commands.................................................................................33
2.11.1 Commands...................................................................................................34
2.11.2 Variable lists...............................................................................................35
2.11.3 if and in qualifiers......................................................................................36
Examples of if qualifier.............................................................................37
2.11.4 Options........................................................................................................37
2.12 Managing data.......................................................................................................37
2.12.1 Looking at your data.................................................................................37
2.12.2 Getting information about variables.........................................................38
2.12.3 Missing values............................................................................................41
2.12.4 Selecting observations..............................................................................41
2.12.5 Selecting variables......................................................................................42
2.13 Creating new variables........................................................................................42
2.13.1 generate command......................................................................................42
2.13.2 replace command........................................................................................44
2.13.3 recode command.........................................................................................44
2.13.4 Common transformations for RHS variables............................................45
Breaking a categorical variable into a set of binary variables............45
More examples of creating binary variables..........................................47
Nonlinear transformations.........................................................................48
Interaction terms........................................................................................49
2.14 Labeling variables and values............................................................................49
2.14.1 Variable labels............................................................................................49
2.14.2 Value labels................................................................................................50
2.14.3 notes command...........................................................................................52
2.15 Global and local macros......................................................................................52
2.16 Graphics.................................................................................................................54
2.16.1 graph command..........................................................................................56
2.16.2 Displaying previously drawn graphs.........................................................63
2.16.3 Printing graphs...........................................................................................63
2.16.4 combining graphs........................................................................................63
2.17 A brief tutorial.....................................................................................................65
A batch version....................................................................................72
3 Estimation, testing, fit, and interpretation...............................................................75
3.1 Estimation...............................................................................................................76
3.1.1 Stata's output for ML estimation..............................................................76
3.1.2 ML and sample size...................................................................................77
3.1.3 Problems in obtaining ML estimates........................................................77
3.1.4 Syntax of estimation commands...............................................................78
Variable lists..............................................................................................78
Specifying the estimation sample..............................................................79
Weights.......................................................................................................84
Options.......................................................................................................85
3.1.5 Reading the output.....................................................................................87
Header.........................................................................................................87
Estimates and standard errors.................................................................88
Confidence intervals..................................................................................88
3.1.6 Storing estimation results..........................................................................89
3.1.7 Reformatting output with estimates table................................................89
3.1.8 Reformatting output with estout...............................................................91
3.1.9 Alternative output with listcoef................................................................94
Options for types of coefficients.............................................................95
Options for mlogit, mprobit, and slogit...................................................95
Other options..............................................................................................96
Standardized coefficients...........................................................................96
Factor and percent change........................................................................98
3.2 Postestimation analysis.........................................................................................99
3.3 Testing....................................................................................................................99
3.3.1 Wald tests.....................................................................................................99
The accumulate option...............................................................................101
3.3.2 LR tests.......................................................................................................101
Avoiding invalid LR tests..........................................................................102
3.4 estat command......................................................................................................103
3.5 Measures of fit......................................................................................................104
Syntax of fitstat.....................................................................................104
Options....................................................................................................105
Models and measures.............................................................................105
Example of fitstat...................................................................................107
Methods and formulas for fitstat..........................................................108
3.6 Interpretation.........................................................................................................113
3.6.1 Approaches to interpretation.....................................................................116
3.6.2 Predictions using predict...........................................................................116
3.6.3 Overview of prvalue, prchange, prtab, and pregen...............................118
Specifying the levels of variables............................................................118
Options controlling output........................................................................119
3.6.4 Syntax for prvalue....................................................................................120
Options.......................................................................................................120
Options for confidence intervals..............................................................120
Options used for bootstrapped confidence intervals.............................121
3.6.5 Syntax for prchange..................................................................................122
Options.......................................................................................................122
3.6.6 Syntax for prtab........................................................................................122
Options........................................................................................................123
3.6.7 Syntax for prgen.......................................................................................123
Options........................................................................................................123
Options for confidence intervals and marginals....................................124
Variables generated....................................................................................124
3.6.8 Computing marginal effects using mfx.....................................................125
3.7 Confidence intervals for predictions...................................................................126
3.8 Next steps.............................................................................................................128
II Models for Specific Kinds of Outcomes.................................................................129
4 Models for binary outcomes.....................................................................................131
4.1 The statistical model............................................................................................132
4.1.1 A latent-variable model.............................................................................132
4.1.2 A nonlinear probability model...................................................................135
4.2 Estimation using logit and probit........................................................................136
Variable lists...............................................................................................136
Specifying the estimation sample.............................................................136
Weights.......................................................................................................136
Options........................................................................................................137
Example......................................................................................................137
4.2.1 Observations predicted perfectly..............................................................140
4.3 Hypothesis testing with test and lrtest.............................................................140
4.3.1 Testing Individual coefficients..................................................................140
One-and two-tailed tests..........................................................................141
Testing single coefficients using test.....................................................142
Testing single coefficients using lrtest...................................................142
4.3.2 Testing multiple coefficients.....................................................................143
Testing multiple coefficients using test..................................................143
Testing multiple coefficients using lrtest...............................................144
4.3.3 Comparing LR and Wald tests..................................................................144
4.4 Residuals and influence using predict...............................................................145
4.4.1 Residuals.....................................................................................................147
Example......................................................................................................147
4.4.2 Influential cases..........................................................................................151
4.4.3 Least likely observations...........................................................................152
Syntax..........................................................................................................152
Options.........................................................................................................152
Options controlling the list of values......................................................153
4.5 Measuring fit.........................................................................................................154
4.5.1 Scalar measures of fit using fitstat..........................................................154
4.5.2 Hosmer-Lemeshow statistic.......................................................................155
4.6 Interpretation using predicted values.................................................................157
4.6.1 Predicted probabilities with predict..........................................................158
4.6.2 Individual predicted probabilities with prvalue........................................160
4.6.3 Tables of predicted probabilities with prtab...........................................162
4.6.4 Graphing predicted probabilities with prgen...........................................163
4.6.5 Plotting confidence intervals....................................................................166
4.6.6 Changes in predicted probabilities..........................................................168
Marginal change........................................................................................168
Discrete change.........................................................................................170
4.7 Interpretation using odds ratios with listcoef....................................................177
Multiplicative coefficients.......................................................................179
Effect of the base probability................................................................179
Percent change in the odds...................................................................180
4.8 Other commands for binary outcomes................................................................181
5 Models for ordinal outcomes...................................................................................183
5.1 The statistical model............................................................................................184
5.1.1 A latent-variable model............................................................................184
5.1.2 A nonlinear probability model...................................................................187
5.2 Estimation using ologit and oprobit....................................................................188
Variable lists...........................................................................................188
Specifying the estimation sample..........................................................188
Weights.....................................................................................................188
Options......................................................................................................189
5.2.1 Example of attitudes toward working mothers........................................189
5.2.2 Predicting perfectly.....................................................................................192
5.3 Hypothesis testing with test and lrtest..............................................................193
5.3.1 Testing individual coefficients...................................................................193
5.3.2 Testing multiple coefficients.....................................................................194
5.4 Scalar measures of fit using fitstat....................................................................195
5.5 Converting to a different parameterization*......................................................196
5.6 The parallel regression assumption....................................................................197
5.7 Residuals and outliers using predict...................................................................200
5.8 Interpretation.........................................................................................................202
5.8.1 Marginal change in y*...............................................................................203
5.8.2 Predicted probabilities...............................................................................204
5.8.3 Predicted probabilities with predict..........................................................204
5.8.4 Individual predicted probabilities with prvalue........................................205
5.8.5 Tables of predicted probabilities with prtab............................................207
5.8.6 Graphing predicted probabilities with prgen............................................209
5.8.7 Changes in predicted probabilities............................................................211
Marginal change with prchange................................................................212
Marginal change with mfx.........................................................................212
Discrete change with prchange.................................................................213
Confidence intervals for discrete changes..............................................215
Computing discrete change for a 10-year increase in age..................216
5.8.8 Odds ratios using listcoef.........................................................................217
5.9 Less common models for ordinal outcomes.....................................................220
5.9.1 The stereotype model...............................................................................220
5.9.2 The generalized ordered logit model......................................................220
5.9.3 The continuation ratio model...................................................................221
6 Models for nominal outcomes with case-specific data........................................223
6.1 The multinomial logit model...............................................................................224
6.1.1 Formal statement of the model................................................................227
6.2 Estimation using mlogit.......................................................................................228
Variable lists............................................................................................228
Specifying the estimation sample...........................................................229
Weights......................................................................................................229
Options.......................................................................................................229
6.2.1 Example of occupational attainment.........................................................230
6.2.2 Using different base categories................................................................231
6.2.3 Predicting perfectly....................................................................................234
6.3 Hypothesis testing of coefficients......................................................................234
6.3.1 mlogtest for tests of the MNLM...............................................................235
Options.........................................................................................................235
6.3.2 Testing the effects of the independent variables...................................236
A likelihood-ratio test...............................................................................236
A Wald test................................................................................................237
Testing multiple independent variables....................................................238
6.3.3 Tests for combining alternatives...............................................................239
A Wald test for combining alternatives...................................................239
Using test [category]*..............................................................................240
An LR test for combining alternatives.....................................................241
Using constraint with lrtest*.....................................................................241
6.4 Independence of irrelevant alternatives.............................................................243
Hausman test of HA...............................................................................243
Small-Hsiao test of HA..........................................................................245
6.5 Measures of fit......................................................................................................246
6.6 Interpretation.........................................................................................................246
6.6.1 Predicted probabilities................................................................................247
6.6.2 Predicted probabilities with predict ........................................................247
Using predict to compare mlogit and ologit............................................248
6.6.3 Predicted probabilities and discrete change with prvalue.....................249
6.6.4 Tables of predicted probabilities with prtab...........................................249
6.6.5 Graphing predicted probabilities with prgen............................................250
Plotting probabilities for one outcome and two groups.........................251
Graphing probabilities for all outcomes for one group..........................252
6.6.6 Changes in predicted probabilities............................................................254
Computing marginal and discrete change with prchange.......................255
Marginal change with mfx.........................................................................257
6.6.7 Plotting discrete changes with prchange and mlogview........................257
6.6.8 Odds ratios using listcoef and mlogview.................................................260
Listing odds ratios with listcoef...............................................................261
Plotting odds ratios....................................................................................262
6.6.9 Using mlogplot*..........................................................................................267
6.6.10 Plotting estimates from matrices with mlogplot*..................................268
Options for using matrices with mlogplot*............................................269
Global macros and matrices used by mlogplot.....................................269
Example.....................................................................................................270
6.7 Multinomial probit model with IIA......................................................................272
6.8 Stereotype logistic regression............................................................................277
6.8.1 Formal statement of the one-dimensional SLM......................................279
6.8.2 Fitting the SLM with slogit........................................................................280
Options.........................................................................................................280
Example.......................................................................................................281
6.8.3 Interpretation using predicted probabilities..............................................281
6.8.4 Interpretation using odds ratios................................................................283
6.8.5 Distinguish ability and the ϕ parameters..................................................286
6.8.6 Ordinality in the one-dimensional SLM....................................................288
Higher-dimension SLM..............................................................................291
7 Models for nominal outcomes with alternative-specific data.................................293
7.1 Alternative-specific data organization...............................................................294
7.1.1 Syntax fro case2alt...................................................................................296
7.2 The conditional logit model.................................................................................297
7.2.1 Fitting the conditional logit model............................................................298
Example of the clogit model....................................................................298
7.2.2 Interpreting odds ratios from clogit.........................................................299
7.2.3 Interpreting probabilities from clogit........................................................299
Using predict..............................................................................................299
Using asprvalue.........................................................................................300
7.2.4 Fitting the multinomial logit model using clogit.....................................304
Setting up the data with case2alt............................................................304
Fitting multinomial logit with clogit.........................................................306
7.2.5 Using clogit with case-and alternative-specific variables....................307
Example of a mixed model......................................................................308
Interpretation of odds ratios using listcoef.............................................308
Interpretation of predicted probabilities using asprvalue......................310
Allowing the effects of alternative-specific variables to vary
over the alternatives.................................................................312
7.3 Alternative-specific multinomial probit..............................................................313
7.3.1 The model...................................................................................................314
7.3.2 Informal explanation of estimation by simulation....................................315
7.3.3 Alternative-based data with uncorrelated errors....................................319
Options........................................................................................................319
Examples.....................................................................................................320
7.3.4 Alternative-based data with correlated errors.......................................322
7.4 The structural covariance matrix.......................................................................325
7.4.1 Interpretation using probabilities...............................................................329
Using predict...............................................................................................329
Using asprvalue..........................................................................................330
7.4.2 Identification, discrete change, and marginal effects.............................332
7.4.3 Testing for IIA...........................................................................................336
7.4.4 Adding case-specific data........................................................................337
7.5 Rank-ordered logistic regression......................................................................339
7.5.1 Fitting the rank-ordered logit model......................................................341
Options.......................................................................................................341
Example of the rank-ordered logit model.............................................342
7.5.2 Interpreting results from rologit..............................................................343
Interpretation using odds ratios...............................................................343
Interpretation using predicted probabilities............................................345
7.6 Conclusions...........................................................................................................347
8 Models for count outcomes.....................................................................................349
8.1 The Poisson distribution......................................................................................349
8.1.1 Fitting the Poisson distribution with the poisson command..................350
8.1.2 Computing predicted probabilities with prcounts....................................352
Syntax.........................................................................................................352
Option..........................................................................................................352
Variables generated...................................................................................352
8.1.3 Comparing observed and predicted counts with prcounts.....................354
8.2 The Poisson regression model...........................................................................356
8.2.1 Fitting the PRM with poisson...................................................................357
Variable lists..............................................................................................357
Specifying the estimation sample............................................................358
Weights.......................................................................................................358
Options.......................................................................................................358
8.2.2 Example of fitting the PRM.....................................................................358
8.2.3 Interpretation using the rate, μ...............................................................359
Factor change in E(y｜x).........................................................................359
Percent change in E(y｜x).......................................................................360
Example of factor and percent change...................................................360
Marginal change in E(y｜x)......................................................................361
Example of marginal change using prchange..........................................362
Example of marginal change using mfx...................................................362
Discrete change in E(y｜x).......................................................................362
Example of discrete change using prchnage..........................................363
Example of discrete change with confidence intervals.........................364
8.2.4 Interpretation using predicted probabilities.............................................365
Example of predicted probabilities using prvalue..................................365
Example of predicted probabilities using prgen.....................................367
Example of predicted probabilities using prcounts................................368
8.2.5 Exposure time*...........................................................................................370
8.3 The negative binomial regression model...........................................................372
8.3.1 Fitting the NBRM with nbreg....................................................................374
NB1and NB2 variance functions...............................................................374
8.3.2 Example of fitting the NBRM....................................................................375
Comparing the PRMand NBRM using estimates table............................375
8.3.3 Testing for overdispersion.........................................................................376
8.3.4 Interpretation using the rate μ..................................................................377
8.3.5 Interpretation using predicted probabilities..............................................378
8.4 Models for truncated counts................................................................................381
8.4.1 Fitting zero-truncated models...................................................................383
8.4.2 Example of fitting zero-truncated models...............................................383
8.4.3 Interpretation of parameters......................................................................384
8.4.4 Interpretation using predicted probabilities and rates............................386
8.4.5 Computing predicted rates and probabilities in the estimation sample...387
8.5 The hurdle regression model*.............................................................................387
8.5.1 In-sample predictions for the hurdle model............................................388
8.5.2 Predictions for user-specified values.......................................................391
8.6 Zero-inflated count models..................................................................................394
8.6.1 Fitting zero-inflated models with zinb and zip........................................396
Variable lists...............................................................................................397
Options.........................................................................................................397
8.6.2 Example of fitting the ZIP and ZINB models..........................................397
8.6.3 Interpretation of coefficients.....................................................................398
8.6.4 Interpretation of predicted probabilities...................................................400
Predicted probabilities with prvalue.........................................................400
Confidence intervals with prvalue............................................................401
Predicted probabilities with prgen...........................................................404
8.7 Comparisons among count models.....................................................................405
8.7.1 Comparing mean probabilities...................................................................405
8.7.2 Tests to compare count models..............................................................407
LR tests of α.............................................................................................407
Vuong test of nonnested models............................................................408
8.8 Using countfit to compare count models.........................................................409
9 More topics..............................................................................................................415
9.1 Ordinal and nominal independent variables.....................................................415
9.1.1 Coding a categorical independent variable as a set of dummy
variables....................................................................................................415
9.1.2 Estimation and interpretation with categorical independent
variables....................................................................................................417
9.1.3 Tests with categorical independent variables.......................................418
Testing the effect of membership in one category versus the
reference category.....................................................................418
Testing the effect of membership in two nonreference categories...419
Testing that a categorical independent variable has no effect..........420
Testing whether treating an ordinal variable as interval loses
information.................................................................................421
9.1.4 Discrete change for categorical independent variables........................422
Computing discrete change with prchange............................................422
Computing discrete change with prvalue...............................................423
9.2 Interactions..........................................................................................................423
9.2.1 Computing sex differences in predictions with interactions................425
9.2.2 Computing sex differences in discrete change with interactions........426
9.3 Nonlinear nonlinear models................................................................................427
9.3.1 Adding nonlinearities to linear predictors..............................................428
9.3.2 Discrete change in nonlinear models......................................................429
9.4 Using praccum and for values to plot predictions..........................................430
Options......................................................................................................431
9.4.1 Example using age and age-squared......................................................432
9.4.2 Using forvalues with praccum..................................................................434
9.4.3 Using praccum for graphing a transformed variable.............................435
9.4.4 Using praccum to graph interactions......................................................436
9.4.5 Using forvalues with prvalue to create tables.......................................438
9.4.6 A more advanced example*.....................................................................441
9.4.7 Using forvalues to create tables with other commands.......................442
9.5 Extending SPost to other estimation commands.............................................444
9.6 Using Stata more efficiently..............................................................................444
9.6.1 profile.do....................................................................................................444
9.6.2 Changing screen fonts and window preferences...................................446
9.6.3 Using ado-files for changing directories................................................446
9.6.4 me.hlp file..................................................................................................446
9.7 Conclusions..........................................................................................................447
A Syntax for SPost commands..................................................................................449
A.1 asprvalue.............................................................................................................450
Syntax...................................................................................................450
Description............................................................................................450
Options..................................................................................................450
Examples...............................................................................................451
A.2 brant.....................................................................................................................452
Syntax...................................................................................................452
Description............................................................................................452
Options..................................................................................................452
Examples..............................................................................................453
Saved results.......................................................................................454
A.3 case2alt................................................................................................................454
Syntax....................................................................................................454
Description............................................................................................454
Options..................................................................................................454
Example.................................................................................................455
A.4 countfit.................................................................................................................455
Syntax....................................................................................................456
Description............................................................................................456
Options for specifying the model........................................................456
Options to select the models to fit....................................................456
Options to label and save results.......................................................456
Options to control what is printed......................................................457
Example..................................................................................................457
A.5 fitstat.....................................................................................................................459
Syntax....................................................................................................459
Description............................................................................................459
Options..................................................................................................459
Examples..............................................................................................459
Saved results.......................................................................................461
A.6 leastlikely............................................................................................................461
Syntax..................................................................................................461
Description..........................................................................................461
Options................................................................................................462
Options for listing..............................................................................462
Examples.............................................................................................462
A.7 listcoef................................................................................................................464
Syntax..................................................................................................464
Description...........................................................................................464
Options.................................................................................................464
Options for nominal outcomes...........................................................465
Examples.............................................................................................465
Saved results......................................................................................467
A.8 misschk..............................................................................................................468
Syntax.................................................................................................468
Options...............................................................................................468
Example.............................................................................................468
A.9 mlogplot............................................................................................................470
Syntax................................................................................................470
Description........................................................................................470
Options..............................................................................................471
Examples..........................................................................................472
A.10 mlogtest.........................................................................................................473
Syntax...............................................................................................473
Description.......................................................................................473
Options..............................................................................................473
Examples..........................................................................................474
Saved results...................................................................................476
Acknowledgment..............................................................................476
A.11 mlogiview.......................................................................................................477
Syntax................................................................................................477
Description........................................................................................477
Dialog box controls...........................................................................477
A.12 Overview of prchange, prgen, prtab, and prvalue......................................478
Syntax................................................................................................478
Examples...........................................................................................479
A.13 praccum...........................................................................................................480
Syntax.................................................................................................480
Description........................................................................................480
Options..............................................................................................480
Examples...........................................................................................481
Variables generated........................................................................482
A.14 prchange........................................................................................................483
Syntax...............................................................................................483
Description........................................................................................483
Options.............................................................................................483
Examples.........................................................................................484
A.15 prcounts........................................................................................................485
Syntax..............................................................................................485
Description......................................................................................485
Options............................................................................................485
Variables generated.......................................................................486
Examples.........................................................................................486
A.16 prgen............................................................................................................487
Syntax.............................................................................................487
Description.....................................................................................487
Options.............................................................................................488
Options for confidence intervals and marginals............................488
Examples...........................................................................................488
Variables generated.........................................................................489
A.17 prtab...............................................................................................................490
Syntax...............................................................................................490
Description.......................................................................................490
Options..............................................................................................491
Examples..........................................................................................491
A.18 prvalue..........................................................................................................493
Syntax...............................................................................................493
Description.......................................................................................493
Options..............................................................................................494
Options for confidence intervals.....................................................494
Options used for bootstrapped confidence intervals.....................494
Examples............................................................................................495
Saved results.....................................................................................497
A.19 spex.................................................................................................................498
Syntax.................................................................................................498
Description.........................................................................................498
Options...............................................................................................498
Examples...........................................................................................498
B Description of datasets........................................................................................498
B.1 binlfp2...............................................................................................................499
B.2 couart2..............................................................................................................499
B.3 gsskidvalue2....................................................................................................501
B.4 nomocc2..........................................................................................................502
B.5 ordwarm2........................................................................................................503
B.6 science2..........................................................................................................504
B.7 travel2.............................................................................................................506
B.8 wlsrnk..............................................................................................................507
References...............................................................................................................509
Author index............................................................................................................515
Subject index...........................................................................................................517
I General Information.....................................................................................................1
1 Introduction..................................................................................................................3
1.1 What is this book about?......................................................................................3
1.2 Which models are considered?.............................................................................4
1.3 Whom is this book for?.........................................................................................5
1.4 How is the book organized?.................................................................................5
1.5 What software do you need?.................................................................................7
1.5.1 Updating Stata 9............................................................................................8
1.5.2 Installing SPost..............................................................................................9
Installing SPost using search.......................................................................9
Installing SPost using net install................................................................11
1.5.3 What if commands do not work?................................................................11
1.5.4 Uninstalling SPost.........................................................................................12
1.5.5 Using spex to load data and run examples...............................................12
1.5.6 More files available on the web site..........................................................13
1.6 Where can I learn more about the models?.........................................................13
2 Introduction to Stata....................................................................................................15
2.1 The Stata interface.................................................................................................16
Changing the scrollback buffer size.......................................................18
Changing the display of variable names in the Variables window ....19
2.2 Abbreviations...........................................................................................................19
2.3 How to get help......................................................................................................20
2.3.1 Online help.....................................................................................................20
2.3.2 Manuals..........................................................................................................21
2.3.3 Other resources.............................................................................................21
2.4 The working directory...........................................................................................21
2.5 Stata file types........................................................................................................22
2.6 Saving output to log files......................................................................................22
Options......................................................................................................23
2.6.1 Closing a log file..........................................................................................23
2.6.2 Viewing a log file........................................................................................24
2.6.3 Converting from SMCL to plain text or PostScript ................................24
2.7 Using and saving datasets....................................................................................24
2.7.1 Data in Stata format....................................................................................24
2.7.2 Data in other formats..................................................................................25
2.7.3 Entering data by hand.................................................................................26
2.8 Size limitations on datasets*................................................................................26
2.9 Do-files...................................................................................................................26
2.9.1 Adding comments.........................................................................................28
2.9.2 Long lines.....................................................................................................29
2.9.3 Stopping a do-file while it is running.......................................................29
2.9.4 Creating do-files..........................................................................................29
Using Stata's Do-file Editor.......................................................................29
Using other editors to create do-files......................................................30
2.9.5 Recommended structure for do-files.........................................................30
2.10 Using Stata for serious data analysis................................................................31
2.11 Syntax of Stata commands.................................................................................33
2.11.1 Commands...................................................................................................34
2.11.2 Variable lists...............................................................................................35
2.11.3 if and in qualifiers......................................................................................36
Examples of if qualifier.............................................................................37
2.11.4 Options........................................................................................................37
2.12 Managing data.......................................................................................................37
2.12.1 Looking at your data.................................................................................37
2.12.2 Getting information about variables.........................................................38
2.12.3 Missing values............................................................................................41
2.12.4 Selecting observations..............................................................................41
2.12.5 Selecting variables......................................................................................42
2.13 Creating new variables........................................................................................42
2.13.1 generate command......................................................................................42
2.13.2 replace command........................................................................................44
2.13.3 recode command.........................................................................................44
2.13.4 Common transformations for RHS variables............................................45
Breaking a categorical variable into a set of binary variables............45
More examples of creating binary variables..........................................47
Nonlinear transformations.........................................................................48
Interaction terms........................................................................................49
2.14 Labeling variables and values............................................................................49
2.14.1 Variable labels............................................................................................49
2.14.2 Value labels................................................................................................50
2.14.3 notes command...........................................................................................52
2.15 Global and local macros......................................................................................52
2.16 Graphics.................................................................................................................54
2.16.1 graph command..........................................................................................56
2.16.2 Displaying previously drawn graphs.........................................................63
2.16.3 Printing graphs...........................................................................................63
2.16.4 combining graphs........................................................................................63
2.17 A brief tutorial.....................................................................................................65
A batch version....................................................................................72
3 Estimation, testing, fit, and interpretation...............................................................75
3.1 Estimation...............................................................................................................76
3.1.1 Stata's output for ML estimation..............................................................76
3.1.2 ML and sample size...................................................................................77
3.1.3 Problems in obtaining ML estimates........................................................77
3.1.4 Syntax of estimation commands...............................................................78
Variable lists..............................................................................................78
Specifying the estimation sample..............................................................79
Weights.......................................................................................................84
Options.......................................................................................................85
3.1.5 Reading the output.....................................................................................87
Header.........................................................................................................87
Estimates and standard errors.................................................................88
Confidence intervals..................................................................................88
3.1.6 Storing estimation results..........................................................................89
3.1.7 Reformatting output with estimates table................................................89
3.1.8 Reformatting output with estout...............................................................91
3.1.9 Alternative output with listcoef................................................................94
Options for types of coefficients.............................................................95
Options for mlogit, mprobit, and slogit...................................................95
Other options..............................................................................................96
Standardized coefficients...........................................................................96
Factor and percent change........................................................................98
3.2 Postestimation analysis.........................................................................................99
3.3 Testing....................................................................................................................99
3.3.1 Wald tests.....................................................................................................99
The accumulate option...............................................................................101
3.3.2 LR tests.......................................................................................................101
Avoiding invalid LR tests..........................................................................102
3.4 estat command......................................................................................................103
3.5 Measures of fit......................................................................................................104
Syntax of fitstat.....................................................................................104
Options....................................................................................................105
Models and measures.............................................................................105
Example of fitstat...................................................................................107
Methods and formulas for fitstat..........................................................108
3.6 Interpretation.........................................................................................................113
3.6.1 Approaches to interpretation.....................................................................116
3.6.2 Predictions using predict...........................................................................116
3.6.3 Overview of prvalue, prchange, prtab, and pregen...............................118
Specifying the levels of variables............................................................118
Options controlling output........................................................................119
3.6.4 Syntax for prvalue....................................................................................120
Options.......................................................................................................120
Options for confidence intervals..............................................................120
Options used for bootstrapped confidence intervals.............................121
3.6.5 Syntax for prchange..................................................................................122
Options.......................................................................................................122
3.6.6 Syntax for prtab........................................................................................122
Options........................................................................................................123
3.6.7 Syntax for prgen.......................................................................................123
Options........................................................................................................123
Options for confidence intervals and marginals....................................124
Variables generated....................................................................................124
3.6.8 Computing marginal effects using mfx.....................................................125
3.7 Confidence intervals for predictions...................................................................126
3.8 Next steps.............................................................................................................128
II Models for Specific Kinds of Outcomes.................................................................129
4 Models for binary outcomes.....................................................................................131
4.1 The statistical model............................................................................................132
4.1.1 A latent-variable model.............................................................................132
4.1.2 A nonlinear probability model...................................................................135
4.2 Estimation using logit and probit........................................................................136
Variable lists...............................................................................................136
Specifying the estimation sample.............................................................136
Weights.......................................................................................................136
Options........................................................................................................137
Example......................................................................................................137
4.2.1 Observations predicted perfectly..............................................................140
4.3 Hypothesis testing with test and lrtest.............................................................140
4.3.1 Testing Individual coefficients..................................................................140
One-and two-tailed tests..........................................................................141
Testing single coefficients using test.....................................................142
Testing single coefficients using lrtest...................................................142
4.3.2 Testing multiple coefficients.....................................................................143
Testing multiple coefficients using test..................................................143
Testing multiple coefficients using lrtest...............................................144
4.3.3 Comparing LR and Wald tests..................................................................144
4.4 Residuals and influence using predict...............................................................145
4.4.1 Residuals.....................................................................................................147
Example......................................................................................................147
4.4.2 Influential cases..........................................................................................151
4.4.3 Least likely observations...........................................................................152
Syntax..........................................................................................................152
Options.........................................................................................................152
Options controlling the list of values......................................................153
4.5 Measuring fit.........................................................................................................154
4.5.1 Scalar measures of fit using fitstat..........................................................154
4.5.2 Hosmer-Lemeshow statistic.......................................................................155
4.6 Interpretation using predicted values.................................................................157
4.6.1 Predicted probabilities with predict..........................................................158
4.6.2 Individual predicted probabilities with prvalue........................................160
4.6.3 Tables of predicted probabilities with prtab...........................................162
4.6.4 Graphing predicted probabilities with prgen...........................................163
4.6.5 Plotting confidence intervals....................................................................166
4.6.6 Changes in predicted probabilities..........................................................168
Marginal change........................................................................................168
Discrete change.........................................................................................170
4.7 Interpretation using odds ratios with listcoef....................................................177
Multiplicative coefficients.......................................................................179
Effect of the base probability................................................................179
Percent change in the odds...................................................................180
4.8 Other commands for binary outcomes................................................................181
5 Models for ordinal outcomes...................................................................................183
5.1 The statistical model............................................................................................184
5.1.1 A latent-variable model............................................................................184
5.1.2 A nonlinear probability model...................................................................187
5.2 Estimation using ologit and oprobit....................................................................188
Variable lists...........................................................................................188
Specifying the estimation sample..........................................................188
Weights.....................................................................................................188
Options......................................................................................................189
5.2.1 Example of attitudes toward working mothers........................................189
5.2.2 Predicting perfectly.....................................................................................192
5.3 Hypothesis testing with test and lrtest..............................................................193
5.3.1 Testing individual coefficients...................................................................193
5.3.2 Testing multiple coefficients.....................................................................194
5.4 Scalar measures of fit using fitstat....................................................................195
5.5 Converting to a different parameterization*......................................................196
5.6 The parallel regression assumption....................................................................197
5.7 Residuals and outliers using predict...................................................................200
5.8 Interpretation.........................................................................................................202
5.8.1 Marginal change in y*...............................................................................203
5.8.2 Predicted probabilities...............................................................................204
5.8.3 Predicted probabilities with predict..........................................................204
5.8.4 Individual predicted probabilities with prvalue........................................205
5.8.5 Tables of predicted probabilities with prtab............................................207
5.8.6 Graphing predicted probabilities with prgen............................................209
5.8.7 Changes in predicted probabilities............................................................211
Marginal change with prchange................................................................212
Marginal change with mfx.........................................................................212
Discrete change with prchange.................................................................213
Confidence intervals for discrete changes..............................................215
Computing discrete change for a 10-year increase in age..................216
5.8.8 Odds ratios using listcoef.........................................................................217
5.9 Less common models for ordinal outcomes.....................................................220
5.9.1 The stereotype model...............................................................................220
5.9.2 The generalized ordered logit model......................................................220
5.9.3 The continuation ratio model...................................................................221
6 Models for nominal outcomes with case-specific data........................................223
6.1 The multinomial logit model...............................................................................224
6.1.1 Formal statement of the model................................................................227
6.2 Estimation using mlogit.......................................................................................228
Variable lists............................................................................................228
Specifying the estimation sample...........................................................229
Weights......................................................................................................229
Options.......................................................................................................229
6.2.1 Example of occupational attainment.........................................................230
6.2.2 Using different base categories................................................................231
6.2.3 Predicting perfectly....................................................................................234
6.3 Hypothesis testing of coefficients......................................................................234
6.3.1 mlogtest for tests of the MNLM...............................................................235
Options.........................................................................................................235
6.3.2 Testing the effects of the independent variables...................................236
A likelihood-ratio test...............................................................................236
A Wald test................................................................................................237
Testing multiple independent variables....................................................238
6.3.3 Tests for combining alternatives...............................................................239
A Wald test for combining alternatives...................................................239
Using test [category]*..............................................................................240
An LR test for combining alternatives.....................................................241
Using constraint with lrtest*.....................................................................241
6.4 Independence of irrelevant alternatives.............................................................243
Hausman test of HA...............................................................................243
Small-Hsiao test of HA..........................................................................245
6.5 Measures of fit......................................................................................................246
6.6 Interpretation.........................................................................................................246
6.6.1 Predicted probabilities................................................................................247
6.6.2 Predicted probabilities with predict ........................................................247
Using predict to compare mlogit and ologit............................................248
6.6.3 Predicted probabilities and discrete change with prvalue.....................249
6.6.4 Tables of predicted probabilities with prtab...........................................249
6.6.5 Graphing predicted probabilities with prgen............................................250
Plotting probabilities for one outcome and two groups.........................251
Graphing probabilities for all outcomes for one group..........................252
6.6.6 Changes in predicted probabilities............................................................254
Computing marginal and discrete change with prchange.......................255
Marginal change with mfx.........................................................................257
6.6.7 Plotting discrete changes with prchange and mlogview........................257
6.6.8 Odds ratios using listcoef and mlogview.................................................260
Listing odds ratios with listcoef...............................................................261
Plotting odds ratios....................................................................................262
6.6.9 Using mlogplot*..........................................................................................267
6.6.10 Plotting estimates from matrices with mlogplot*..................................268
Options for using matrices with mlogplot*............................................269
Global macros and matrices used by mlogplot.....................................269
Example.....................................................................................................270
6.7 Multinomial probit model with IIA......................................................................272
6.8 Stereotype logistic regression............................................................................277
6.8.1 Formal statement of the one-dimensional SLM......................................279
6.8.2 Fitting the SLM with slogit........................................................................280
Options.........................................................................................................280
Example.......................................................................................................281
6.8.3 Interpretation using predicted probabilities..............................................281
6.8.4 Interpretation using odds ratios................................................................283
6.8.5 Distinguish ability and the ϕ parameters..................................................286
6.8.6 Ordinality in the one-dimensional SLM....................................................288
Higher-dimension SLM..............................................................................291
7 Models for nominal outcomes with alternative-specific data.................................293
7.1 Alternative-specific data organization...............................................................294
7.1.1 Syntax fro case2alt...................................................................................296
7.2 The conditional logit model.................................................................................297
7.2.1 Fitting the conditional logit model............................................................298
Example of the clogit model....................................................................298
7.2.2 Interpreting odds ratios from clogit.........................................................299
7.2.3 Interpreting probabilities from clogit........................................................299
Using predict..............................................................................................299
Using asprvalue.........................................................................................300
7.2.4 Fitting the multinomial logit model using clogit.....................................304
Setting up the data with case2alt............................................................304
Fitting multinomial logit with clogit.........................................................306
7.2.5 Using clogit with case-and alternative-specific variables....................307
Example of a mixed model......................................................................308
Interpretation of odds ratios using listcoef.............................................308
Interpretation of predicted probabilities using asprvalue......................310
Allowing the effects of alternative-specific variables to vary
over the alternatives.................................................................312
7.3 Alternative-specific multinomial probit..............................................................313
7.3.1 The model...................................................................................................314
7.3.2 Informal explanation of estimation by simulation....................................315
7.3.3 Alternative-based data with uncorrelated errors....................................319
Options........................................................................................................319
Examples.....................................................................................................320
7.3.4 Alternative-based data with correlated errors.......................................322
7.4 The structural covariance matrix.......................................................................325
7.4.1 Interpretation using probabilities...............................................................329
Using predict...............................................................................................329
Using asprvalue..........................................................................................330
7.4.2 Identification, discrete change, and marginal effects.............................332
7.4.3 Testing for IIA...........................................................................................336
7.4.4 Adding case-specific data........................................................................337
7.5 Rank-ordered logistic regression......................................................................339
7.5.1 Fitting the rank-ordered logit model......................................................341
Options.......................................................................................................341
Example of the rank-ordered logit model.............................................342
7.5.2 Interpreting results from rologit..............................................................343
Interpretation using odds ratios...............................................................343
Interpretation using predicted probabilities............................................345
7.6 Conclusions...........................................................................................................347
8 Models for count outcomes.....................................................................................349
8.1 The Poisson distribution......................................................................................349
8.1.1 Fitting the Poisson distribution with the poisson command..................350
8.1.2 Computing predicted probabilities with prcounts....................................352
Syntax.........................................................................................................352
Option..........................................................................................................352
Variables generated...................................................................................352
8.1.3 Comparing observed and predicted counts with prcounts.....................354
8.2 The Poisson regression model...........................................................................356
8.2.1 Fitting the PRM with poisson...................................................................357
Variable lists..............................................................................................357
Specifying the estimation sample............................................................358
Weights.......................................................................................................358
Options.......................................................................................................358
8.2.2 Example of fitting the PRM.....................................................................358
8.2.3 Interpretation using the rate, μ...............................................................359
Factor change in E(y｜x).........................................................................359
Percent change in E(y｜x).......................................................................360
Example of factor and percent change...................................................360
Marginal change in E(y｜x)......................................................................361
Example of marginal change using prchange..........................................362
Example of marginal change using mfx...................................................362
Discrete change in E(y｜x).......................................................................362
Example of discrete change using prchnage..........................................363
Example of discrete change with confidence intervals.........................364
8.2.4 Interpretation using predicted probabilities.............................................365
Example of predicted probabilities using prvalue..................................365
Example of predicted probabilities using prgen.....................................367
Example of predicted probabilities using prcounts................................368
8.2.5 Exposure time*...........................................................................................370
8.3 The negative binomial regression model...........................................................372
8.3.1 Fitting the NBRM with nbreg....................................................................374
NB1and NB2 variance functions...............................................................374
8.3.2 Example of fitting the NBRM....................................................................375
Comparing the PRMand NBRM using estimates table............................375
8.3.3 Testing for overdispersion.........................................................................376
8.3.4 Interpretation using the rate μ..................................................................377
8.3.5 Interpretation using predicted probabilities..............................................378
8.4 Models for truncated counts................................................................................381
8.4.1 Fitting zero-truncated models...................................................................383
8.4.2 Example of fitting zero-truncated models...............................................383
8.4.3 Interpretation of parameters......................................................................384
8.4.4 Interpretation using predicted probabilities and rates............................386
8.4.5 Computing predicted rates and probabilities in the estimation sample...387
8.5 The hurdle regression model*.............................................................................387
8.5.1 In-sample predictions for the hurdle model............................................388
8.5.2 Predictions for user-specified values.......................................................391
8.6 Zero-inflated count models..................................................................................394
8.6.1 Fitting zero-inflated models with zinb and zip........................................396
Variable lists...............................................................................................397
Options.........................................................................................................397
8.6.2 Example of fitting the ZIP and ZINB models..........................................397
8.6.3 Interpretation of coefficients.....................................................................398
8.6.4 Interpretation of predicted probabilities...................................................400
Predicted probabilities with prvalue.........................................................400
Confidence intervals with prvalue............................................................401
Predicted probabilities with prgen...........................................................404
8.7 Comparisons among count models.....................................................................405
8.7.1 Comparing mean probabilities...................................................................405
8.7.2 Tests to compare count models..............................................................407
LR tests of α.............................................................................................407
Vuong test of nonnested models............................................................408
8.8 Using countfit to compare count models.........................................................409
9 More topics..............................................................................................................415
9.1 Ordinal and nominal independent variables.....................................................415
9.1.1 Coding a categorical independent variable as a set of dummy
variables....................................................................................................415
9.1.2 Estimation and interpretation with categorical independent
variables....................................................................................................417
9.1.3 Tests with categorical independent variables.......................................418
Testing the effect of membership in one category versus the
reference category.....................................................................418
Testing the effect of membership in two nonreference categories...419
Testing that a categorical independent variable has no effect..........420
Testing whether treating an ordinal variable as interval loses
information.................................................................................421
9.1.4 Discrete change for categorical independent variables........................422
Computing discrete change with prchange............................................422
Computing discrete change with prvalue...............................................423
9.2 Interactions..........................................................................................................423
9.2.1 Computing sex differences in predictions with interactions................425
9.2.2 Computing sex differences in discrete change with interactions........426
9.3 Nonlinear nonlinear models................................................................................427
9.3.1 Adding nonlinearities to linear predictors..............................................428
9.3.2 Discrete change in nonlinear models......................................................429
9.4 Using praccum and for values to plot predictions..........................................430
Options......................................................................................................431
9.4.1 Example using age and age-squared......................................................432
9.4.2 Using forvalues with praccum..................................................................434
9.4.3 Using praccum for graphing a transformed variable.............................435
9.4.4 Using praccum to graph interactions......................................................436
9.4.5 Using forvalues with prvalue to create tables.......................................438
9.4.6 A more advanced example*.....................................................................441
9.4.7 Using forvalues to create tables with other commands.......................442
9.5 Extending SPost to other estimation commands.............................................444
9.6 Using Stata more efficiently..............................................................................444
9.6.1 profile.do....................................................................................................444
9.6.2 Changing screen fonts and window preferences...................................446
9.6.3 Using ado-files for changing directories................................................446
9.6.4 me.hlp file..................................................................................................446
9.7 Conclusions..........................................................................................................447
A Syntax for SPost commands..................................................................................449
A.1 asprvalue.............................................................................................................450
Syntax...................................................................................................450
Description............................................................................................450
Options..................................................................................................450
Examples...............................................................................................451
A.2 brant.....................................................................................................................452
Syntax...................................................................................................452
Description............................................................................................452
Options..................................................................................................452
Examples..............................................................................................453
Saved results.......................................................................................454
A.3 case2alt................................................................................................................454
Syntax....................................................................................................454
Description............................................................................................454
Options..................................................................................................454
Example.................................................................................................455
A.4 countfit.................................................................................................................455
Syntax....................................................................................................456
Description............................................................................................456
Options for specifying the model........................................................456
Options to select the models to fit....................................................456
Options to label and save results.......................................................456
Options to control what is printed......................................................457
Example..................................................................................................457
A.5 fitstat.....................................................................................................................459
Syntax....................................................................................................459
Description............................................................................................459
Options..................................................................................................459
Examples..............................................................................................459
Saved results.......................................................................................461
A.6 leastlikely............................................................................................................461
Syntax..................................................................................................461
Description..........................................................................................461
Options................................................................................................462
Options for listing..............................................................................462
Examples.............................................................................................462
A.7 listcoef................................................................................................................464
Syntax..................................................................................................464
Description...........................................................................................464
Options.................................................................................................464
Options for nominal outcomes...........................................................465
Examples.............................................................................................465
Saved results......................................................................................467
A.8 misschk..............................................................................................................468
Syntax.................................................................................................468
Options...............................................................................................468
Example.............................................................................................468
A.9 mlogplot............................................................................................................470
Syntax................................................................................................470
Description........................................................................................470
Options..............................................................................................471
Examples..........................................................................................472
A.10 mlogtest.........................................................................................................473
Syntax...............................................................................................473
Description.......................................................................................473
Options..............................................................................................473
Examples..........................................................................................474
Saved results...................................................................................476
Acknowledgment..............................................................................476
A.11 mlogiview.......................................................................................................477
Syntax................................................................................................477
Description........................................................................................477
Dialog box controls...........................................................................477
A.12 Overview of prchange, prgen, prtab, and prvalue......................................478
Syntax................................................................................................478
Examples...........................................................................................479
A.13 praccum...........................................................................................................480
Syntax.................................................................................................480
Description........................................................................................480
Options..............................................................................................480
Examples...........................................................................................481
Variables generated........................................................................482
A.14 prchange........................................................................................................483
Syntax...............................................................................................483
Description........................................................................................483
Options.............................................................................................483
Examples.........................................................................................484
A.15 prcounts........................................................................................................485
Syntax..............................................................................................485
Description......................................................................................485
Options............................................................................................485
Variables generated.......................................................................486
Examples.........................................................................................486
A.16 prgen............................................................................................................487
Syntax.............................................................................................487
Description.....................................................................................487
Options.............................................................................................488
Options for confidence intervals and marginals............................488
Examples...........................................................................................488
Variables generated.........................................................................489
A.17 prtab...............................................................................................................490
Syntax...............................................................................................490
Description.......................................................................................490
Options..............................................................................................491
Examples..........................................................................................491
A.18 prvalue..........................................................................................................493
Syntax...............................................................................................493
Description.......................................................................................493
Options..............................................................................................494
Options for confidence intervals.....................................................494
Options used for bootstrapped confidence intervals.....................494
Examples............................................................................................495
Saved results.....................................................................................497
A.19 spex.................................................................................................................498
Syntax.................................................................................................498
Description.........................................................................................498
Options...............................................................................................498
Examples...........................................................................................498
B Description of datasets........................................................................................498
B.1 binlfp2...............................................................................................................499
B.2 couart2..............................................................................................................499
B.3 gsskidvalue2....................................................................................................501
B.4 nomocc2..........................................................................................................502
B.5 ordwarm2........................................................................................................503
B.6 science2..........................................................................................................504
B.7 travel2.............................................................................................................506
B.8 wlsrnk..............................................................................................................507
References...............................................................................................................509
Author index............................................................................................................515
Subject index...........................................................................................................517