Identification and Inference for Econometric Models: Essays in Honor of Thomas Rothenberg

This document was uploaded by one of our users. The uploader already confirmed that they had the permission to publish it. If you are author/publisher or own the copyright of this documents, please report to us by using this DMCA report form.

Simply click on the Download Book button.

Yes, Book downloads on Ebookily are 100% Free.

Sometimes the book is free on Amazon As well, so go ahead and hit "Search on Amazon"

Author(s): Donald W. K. Andrews, James H. Stock
Year: 2005

Language: English
Pages: 588

Cover......Page 1
Half-title......Page 3
Title......Page 5
Copyright......Page 6
Contents......Page 7
List of Contributors......Page 9
Preface......Page 11
Identification and Efficient Estimation (Part I)......Page 12
Asymptotic Approximations (Part II)......Page 13
Inference Involving Potentially Nonstationary Time Series (Part III)......Page 14
Nonparametric and Semiparametric Inference (Part IV)......Page 15
PART I IDENTIFICATION AND EFFICIENT ESTIMATION......Page 17
1. INTRODUCTION......Page 19
2. EXAMPLE ONE: A MEASUREMENT PROBLEM......Page 20
3. EXAMPLE TWO: A REGRESSION PROBLEM......Page 23
4. IMPLICATIONS FOR ECONOMETRICS......Page 25
References......Page 26
1. INTRODUCTION......Page 27
2. PRIMARY MODEL......Page 28
3. GENETIC THEORY......Page 31
5. AGNOSTIC MODEL......Page 32
7. IDENTIFICATION AND CONSTRAINTS......Page 33
8. EMPIRICAL IMPLEMENTATION......Page 34
9. PRETESTING ISSUES......Page 36
10. MULTIVARIATE MODELS......Page 37
11. OBJECTIVES......Page 38
APPENDIX......Page 39
References......Page 40
1. INTRODUCTION......Page 43
2. THE FRAMEWORK AND SOME BASIC RESULTS......Page 45
3.1. Random Coefficient Model......Page 51
3.2. Probit Model......Page 52
4. APPLICATION TO MODELS WITH ENDOGENOUS EXPLANATORY VARIABLES......Page 59
4.1. Random Coefficient Model......Page 60
4.2. Probit Response Function......Page 62
5. CONCLUSIONS, CAVEATS, AND FURTHER CONSIDERATIONS......Page 66
References......Page 70
1. INTRODUCTION......Page 72
2. A FIRST EXAMPLE: SIMPLE REGRESSION......Page 73
3. CONDITIONALS......Page 75
4. A SECOND EXAMPLE: TWO LINEAR REGRESSIONS......Page 76
5. SIMULTANEOUS EQUATIONS......Page 78
6. NONLINEAR MODELS: FIGURE 4.1 REVISITED......Page 80
7. TECHNICAL NOTES......Page 81
8. MORE COMPLICATED EXAMPLES......Page 82
10. CONCOMITANTS......Page 85
11. THE STORY BEHIND FIGURES 4.3 AND 4.4......Page 86
12. MODELS AND KERNELS REVISITED......Page 88
13. LITERATURE REVIEW......Page 89
References......Page 91
1. INTRODUCTION......Page 96
2.2. k-Class Estimators and Wald Statistics......Page 99
2.4. Weak Instrument Asymptotics: Assumptions and Notation......Page 100
2.5. Selected Weak Instrument Asymptotic Representations......Page 101
3. WEAK INSTRUMENT SETS......Page 102
3.1. First Characterization of a Weak Instrument Set: Bias......Page 103
3.2. Second Characterization of a Weak Instrument Set: Size......Page 104
3.3.1. Weak Instrument Set Based on TSLS bias......Page 105
3.3.3. Weak Instrument Set Based on TSLS Size......Page 106
3.4. Weak Instrument Sets for Other k-Class Estimators......Page 107
3.4.2. BTSLS......Page 108
3.5. Numerical Results for TSLS, LIML, and Fuller-k......Page 109
4. TEST FOR WEAK INSTRUMENTS......Page 111
4.1. A Bound on the Asymptotic Distribution of g min......Page 112
4.3. Critical Values......Page 115
4.3.1. Comparison to the Staiger–Stock Rule of Thumb......Page 117
Interpretation as a Decision Rule......Page 120
6. CONCLUSIONS......Page 121
References......Page 123
1. INTRODUCTION......Page 125
2.1. Model and Notation......Page 126
2.2. k-Class Statistics......Page 127
3. UNIFORM CONVERGENCE RESULT......Page 128
4. MANY WEAK INSTRUMENT ASYMPTOTIC LIMITS......Page 130
ACKNOWLEDGMENTS......Page 131
APPENDIX......Page 132
References......Page 136
1. INTRODUCTION......Page 137
2. IDENTIFICATION IN A MICROSTRUCTURE MODEL WITH OPTIONS MARKETS......Page 138
3. INTRA-TRADING DAY DYNAMICS......Page 143
4. CALENDAR PERIOD IMPLICATIONS......Page 150
6. ACKNOWLEDGMENTS......Page 157
APPENDIX......Page 158
References......Page 160
1. INTRODUCTION......Page 165
2. THE MODEL AND ESTIMATOR......Page 167
3. MAIN RESULTS......Page 169
4. SOME NUMERICAL RESULTS......Page 174
5. CONCLUSIONS......Page 176
APPENDIX......Page 178
References......Page 185
1. INTRODUCTION......Page 187
2. MARKOV MODEL AND MAXIMUM LIKELIHOOD ESTIMATOR......Page 192
3. PARAMETRIC BOOTSTRAP......Page 194
4. ASSUMPTIONS......Page 196
6. HIGHER-ORDER IMPROVEMENTS OF THE PARAMETRIC BOOTSTRAP......Page 198
7. k-STEP PARAMETRIC BOOTSTRAP......Page 199
8. MONTE CARLO SIMULATIONS......Page 204
8.1. Experimental Design......Page 205
8.2. Simulation Results......Page 208
APPENDIX OF PROOFS......Page 213
Lemmas......Page 214
Proof of Theorem 5.1 .......Page 217
Proof of Lemma A.1......Page 223
Proof of Lemma A.2......Page 224
Proof of Lemma A.5......Page 225
Proof of Lemma A.6......Page 226
Proof of Lemma A.7......Page 227
Proof of Lemma A.8......Page 228
References......Page 229
1. INTRODUCTION......Page 232
2. FRAMEWORK......Page 233
3. TWO-STEP GMM ESTIMATOR......Page 235
4. GENERALIZED EMPIRICAL LIKELIHOOD ESTIMATORS......Page 237
5. EMPIRICAL DISCREPANCY THEORY......Page 240
6. A FURTHER CHARACTERIZATION OF ED/GEL ESTIMATORS......Page 242
7. A DETAILED ANALYSIS OF SOME SIMPLE EXAMPLES......Page 244
8. SUMMARY......Page 249
APPENDIX A......Page 252
References......Page 259
1. INTRODUCTION......Page 261
2.2. Sample Structure......Page 263
2.3. GMM and GEL Estimation of alpha 0......Page 264
2.4. GMM and GEL Estimation of Beta 0......Page 265
3.1. The Asymptotic Bias of the Nuisance Parameter Estimator......Page 266
3.2. Independent Samples......Page 267
3.3. Identical Samples......Page 271
4. SIMULATION EXPERIMENTS FOR COVARIANCE STRUCTURE MODELS......Page 274
4.1. Bootstrap Bias Adjustment......Page 275
4.3. Experimental Design......Page 276
4.4. Results......Page 278
5. CONCLUSIONS......Page 283
APPENDIX A: PROOFS......Page 285
APPENDIX B: SOME NOTATION......Page 294
B.4. Asymptotic Bias System-Beta......Page 295
References......Page 296
1. INTRODUCTION......Page 298
2.1. Traditional Instrument-Based Estimators......Page 300
2.2.1. Estimation......Page 301
2.2.2. Inference......Page 303
2.3.1. Moment Validity Tests......Page 304
2.3.2. Tests of Parameter Restrictions......Page 305
2.4. Computational Issues and Approach......Page 306
3.2. Sample Characteristics and Outcome Basis......Page 308
4.1. Estimator MSE Performance......Page 310
4.4. Size of Moment Validity Tests......Page 312
4.6. Test Power......Page 316
5. SOME FINAL REMARKS......Page 318
References......Page 320
1.1. The Model......Page 322
1.2. Realized variance......Page 323
1.3. Mixed Normal Asymptotic Theory......Page 326
2.1. Simple Model......Page 329
2.2. Superposition......Page 331
2.3. Diffusion Case......Page 334
2.4. Alternative Estimators of Quarticity......Page 337
3.2. Relationship between Integrals and Sums......Page 340
3.3. Finite Sample Corrections......Page 342
3.4. Alternative Transformations......Page 343
4. CONCLUSION......Page 344
References......Page 345
1. INTRODUCTION......Page 348
2. GMM ESTIMATORS......Page 350
3. GMM-BASED TESTS......Page 353
4. BOOTSTRAP CRITICAL VALUES......Page 354
5. DESIGN OF EXPERIMENTS AND COMPUTATIONS......Page 356
Asymptotic Critical Values......Page 358
Bootstrap Critical Values: Residuals......Page 363
7. FTW ESTIMATOR......Page 365
8. CONCLUDING COMMENTS......Page 368
ACKNOWLEDGEMENTS......Page 369
References......Page 370
PART III INFERENCE INVOLVING POTENTIALLY NONSTATIONARY TIME SERIES......Page 371
1. INTRODUCTION......Page 373
2. MOTIVATION......Page 374
3. THE MODEL AND ASSUMPTIONS......Page 375
4. A FAMILY OF COINTEGRATION TESTS......Page 377
5. ASYMPTOTIC THEORY......Page 379
APPENDIX A: PROOF OF THEOREM 5.1......Page 383
APPENDIX B: LIMITING DISTRIBUTIONS OF RT AND ST......Page 386
APPENDIX C: PROOF OF THEOREM 5.2......Page 387
References......Page 389
1. INTRODUCTION......Page 391
2. INVERTING A SEQUENCE OF TESTS......Page 392
3. ASYMPTOTIC ANALYSIS......Page 397
3.1. Obtaining Rejection Regions......Page 399
3.2. Asymptotic Interval Length......Page 402
4. MONTE CARLO EVIDENCE......Page 405
Proof of Theorem 3.1 for M Functions......Page 410
References......Page 417
1. INTRODUCTION......Page 419
2. THE FAMILY OF CRAMÉR–VON MISES DISTRIBUTIONS......Page 420
3.1. Testing against the Presence of a Random Walk Component......Page 421
3.2. Serial Correlation......Page 422
3.3. Testing against Nonstationary Seasonality......Page 423
4. UNIT ROOT TESTS......Page 425
4.1. Test Statistics with a Cramér–von Mises Distribution......Page 426
4.2. Serial Correlation and Unobserved Components......Page 428
Stochastic Volatility......Page 429
4.3. Seasonal Unit Root Tests......Page 431
4.4. Slope Unit Root Test......Page 432
5.1. Testing against a Multivariate Random Walk......Page 433
5.2. Multivariate Unit Root Tests......Page 434
6. TESTS WHEN BREAKS ARE PRESENT......Page 435
6.2. Unit Root Tests......Page 436
6.3. Multivariate Series and Seasonality......Page 437
APPENDIX......Page 438
References......Page 439
1. INTRODUCTION......Page 442
2. PANEL STATIONARITY TESTS......Page 445
2.1. The Intercept Only Case: p = 0......Page 446
2.2. The Case with a Linear Trend: p = 1......Page 447
3. A PANEL UNIT ROOT TEST......Page 449
3.1. Monte Carlo Simulations......Page 451
4. APPLICATION TO PPP......Page 459
5. CONCLUSION......Page 461
APPENDIX......Page 462
References......Page 465
1. INTRODUCTION......Page 467
2. TESTING FOR UNIT ROOTS IN PANEL DATA: AN OVERVIEW......Page 470
2.1. Maximum Likelihood Methods with Homoskedastic Errors......Page 471
2.2. Maximum Likelihood Methods with Heteroskedastic Errors......Page 473
2.4. OLS-pooled estimation under the null......Page 474
2.5. The IPS Method......Page 475
3. DESIGN AND CALIBRATION OF SIMULATIONS......Page 476
4. RESULTS OF SIMULATIONS......Page 478
5. RESULTS OF UNIT ROOT TESTS FOR OBSERVED DATA......Page 486
6. CONCLUSIONS......Page 489
APPENDIX A: CML ESTIMATION WITH HETEROSKEDASTICITY......Page 490
APPENDIX B: AUTOCORRELOGRAMS OF THE DATA......Page 492
References......Page 495
1. INTRODUCTION......Page 496
2. BACKGROUND......Page 498
2.1. Terminology......Page 500
2.2. Forecasting and Policy Analysis across Regime Shifts......Page 501
3. THE DATA GENERATION PROCESS......Page 502
3.1. Forecasting Models......Page 504
4. THE IMPACTS OF BREAKS ON THE VEqCM......Page 505
4.1. Breaks in Cointegration Relations......Page 506
4.2. Postbreak Forecasts......Page 507
5. THE IMPACTS OF BREAKS ON THE FORECASTING MODEL......Page 508
6. POLICY REGIME CHANGES......Page 509
7. POLICY CHANGE CORRECTIONS TO ROBUST FORECASTS......Page 510
7.1. Pooling Forecasts......Page 512
7.2. Intercept Corrections......Page 513
7.4. Example......Page 514
8.CONCLUSION......Page 515
References......Page 516
PART IV NONPARAMETRIC AND SEMIPARAMETRIC INFERENCE......Page 519
1. INTRODUCTION......Page 521
2. PRELIMINARIES......Page 523
3. MAIN RESULT......Page 524
4. CRITICAL VALUES AND SIMULATIONS......Page 530
5. DISCUSSION......Page 532
APPENDIX......Page 533
References......Page 534
1. INTRODUCTION......Page 536
2. MOTIVATION FOR THE PROPOSED ESTIMATORS......Page 537
2.1. Partially Linear Logit Model......Page 538
2.2. Partially Linear Tobit Models......Page 539
2.3. Partially Linear Poisson Regression Models......Page 540
2.5. Tobit Models with Selection......Page 541
3. ASYMPTOTIC PROPERTIES OF ESTIMATORS DEFINED BY MINIMIZING KERNEL-WEIGHTED U–STATISTICS......Page 542
3.1. Consistency......Page 543
3.1.2. Pointwise Convergence to Limiting Objective Function......Page 544
3.1.3. Uniform Convergence to Limiting Objective Function......Page 545
3.1.4. Identification......Page 546
3.2. Asymptotic Normality......Page 547
3.2.1. Verifying some of the conditions......Page 551
3.3. Bias Reduction......Page 553
3.3.1. Estimation of the Asymptotic Variance......Page 555
3.4. Asymptotic Properties of Partially Linear Logit Estimator......Page 556
4. MONTE CARLO RESULTS......Page 558
5. POSSIBLE EXTENSIONS......Page 563
APPENDIX: THE MOST DULL DERIVATIONS......Page 565
References......Page 568
1. INTRODUCTION......Page 570
2. THE ESTIMATOR......Page 571
3. ASYMPTOTIC VARIANCE ESTIMATION......Page 574
4. ASYMPTOTIC THEORY......Page 576
5. ASYMPTOTIC EFFICIENCY......Page 578
6. MONTE CARLO EXPERIMENTS......Page 579
APPENDIX: PROOFS......Page 584
References......Page 589