Analysis of Panels and Limited Dependent Variable Models

Cover......Page 1
Half-title......Page 3
Title......Page 5
Copyright......Page 6
Contents......Page 7
Contributors......Page 9
Foreword......Page 11
Introduction......Page 13
References......Page 18
1 Introduction......Page 19
2 A single state model......Page 20
3 Why divide by P......Page 22
Maximizing full LF......Page 24
6 Method which does not require starting-time distribution......Page 25
7 Semiparametric estimation of h(x) and Theta......Page 28
8 Separate estimation of h(x)......Page 29
9 Two-states model......Page 30
Observe both states......Page 31
Observe one state......Page 32
References......Page 33
1 Introduction......Page 35
2 The model......Page 36
3.1 Estimating the reduced form......Page 40
3.2 Asymptotic least squares estimation......Page 42
3.3 Estimates based on orthogonal deviations......Page 43
3.4 Testing the overidentifying restrictions......Page 44
3.5 Consistent OLS estimation using predicted differences......Page 45
4 An application to female labor supply and wages......Page 46
5 Concluding remarks......Page 53
Appendix A Descriptive statistics and additional parameter estimates......Page 54
Appendix B Type I Tobit with symmetric trimming......Page 58
References......Page 59
1 Introduction......Page 61
2.1 Conventional probit model......Page 63
2.2 Mixture of normals probit model......Page 64
2.3 A posterior simulator......Page 65
2.4 Comparison of models......Page 67
3 Some results with artificial data......Page 68
4 An example: labor force participation of women......Page 78
5 Conclusion......Page 88
References......Page 89
1 Introduction......Page 91
2.1 One-limit and two-limit LDRE models......Page 93
2.2 Likelihood functions......Page 95
3 RE solution and simulation......Page 96
4 Likelihood simulation and recursion......Page 100
5 Renewal and variance reduction......Page 102
6 Monte Carlo experiments and results......Page 105
7 Conclusions......Page 120
Appendix: Existence and uniqueness of SRE solution......Page 121
References......Page 123
1 Introduction......Page 126
2 The EC-EM and large T approximation setup......Page 129
3 The Monte Carlo probits experimental design......Page 131
4.1 Design I: Unpoolable slopes......Page 133
4.1 Design II: Poolable slopes......Page 135
5 Concluding remarks......Page 141
References......Page 146
1 Introduction......Page 148
2 Recent empirical investigations of convergence and the rate of convergence......Page 151
3 Alternative methods for estimation......Page 155
3.1 Inconsistency of the pooled-sample OLS estimates of the dynamic error components model......Page 156
3.2 Inconsistency of the OLS estimators of the dummy variable, or fixed-effects, model......Page 157
3.3 Generalized least squares and feasible GLS......Page 160
3.4 Bounds for the coefficient of the lagged dependent variable......Page 161
3.5 Maximum likelihood conditional on the initial value of the lagged dependent variable......Page 162
3.6 Unconditional likelihood and unconditional maximum likelihood......Page 164
4 Empirical evidence on the comparative performance of different panel data methods......Page 168
5 Conclusions......Page 178
References......Page 179
1 Introduction......Page 183
2.1 The model and instrumental variables estimation......Page 185
2.2 Redundancy in instrumental variables......Page 188
3.1 MGIV and GIV......Page 190
3.2 Redundancy results......Page 191
4 Application to models with time-invariant effects......Page 193
4.1 Fixed-effects models......Page 194
4.2 Random-effects models......Page 196
4.3 Hausman and Taylor-type models......Page 197
5 Application to models with time-varying effects......Page 199
6 Monte Carlo experiments......Page 201
7 Conclusions......Page 206
Appendix......Page 207
References......Page 209
1 Introduction......Page 211
2 The stochastic structure of the model......Page 213
3 The bias of the LSDV estimator......Page 217
4 The location of particular IV estimators......Page 221
5 Conclusions......Page 227
Appendix......Page 228
References......Page 236
1 Introduction......Page 238
2 The econometric framework......Page 240
3 Forecast errors and their covariances......Page 242
4 Data and estimates of the error components......Page 249
5 Preliminary versus revised data......Page 254
6 GMM tests for bias......Page 257
7 Martingale test for efficiency......Page 261
8 Conclusion......Page 264
References......Page 265
1 Introduction......Page 267
2 Asymptotic mean squared error (AMSE) of prediction......Page 270
2.1 The ordinary predictor with estimated parameters......Page 271
2.3 The misspecified OLS predictor......Page 273
3 Monte Carlo results......Page 274
4 Conclusion......Page 276
References......Page 278
1 Introduction......Page 280
3 Classical approach......Page 282
4 Bayesian approach......Page 284
5 Asymptotics......Page 289
6 The design of the Monte Carlo study......Page 293
7 Monte Carlo results......Page 295
8 An empirical application: the q investment model re-examined......Page 303
9 Conclusion......Page 305
References......Page 307
1 Introduction......Page 309
2 Dynamic models of heterogeneous panels......Page 311
3 Bias-reduction techniques for estimation of the long-run coefficients......Page 312
3.2 Bias corrections applied directly to the estimator of the long-run coefficients......Page 314
3.3 Bootstrap bias-corrected estimator......Page 317
4.1 Monte Carlo results for a single time series regression......Page 319
4.2 Monte Carlo results for panels......Page 325
5 Concluding remarks......Page 330
References......Page 332
Other research appointments......Page 335
Production functions and productivity......Page 336
Panel data......Page 337
Qualitative variables models......Page 338
Limited dependent variable models......Page 339
Disequilibrium models......Page 340
Time series models......Page 341
Errors in variables......Page 342
Survey data......Page 343
Books......Page 344
Doctoral dissertations......Page 345
Index......Page 346