文档介绍：Chuppter 24
ECONOMETRIC ANALYSIS OF QUALITATIVE
RESPONSE MODELS
DANIEL L. MCFADDEN
Massachusetts Institute of Technology
Contents
1. The problem 1396
2. Binomial response models 1396
. Latent variable specification 1396
. Functional forms 1397
. Estimation 1398
. Contingency table analysis 1400
. Minimum &i-square method 1400
. Discriminant analysis 1401
3. Multinomial response models 1403
. Foundations 1403
. Statistical analysis 1406
. Functional form 1410
. The multinomial logit model 1411
. Independence from irrelevant alternatives 1413
. Limiting the number of alternatives 1415
. Specification tests for the MNL model 1417
. Multinomial probit 1418
. Elimination models 1420
. Hierarchical response models 1422
. An empirical example 1428
4. Further topics 1433
. Extensions 1433
. Dynamic models 1433
. Discrete-continuous systems 1434
. Self-selection and biased samples 1436
. Statistical methods 1439
5. Conclusion 1442
Appendix: Proof outlines for Theorems l-3 1442
References 1446
Handbook of Econometrics, Volume II, Edited by Z. Griliches and . Intriligator
0 Elseoier Science Publishers BV, 1984
1396 D. L. McFadden
1. The problem
An
example is potential bias in analysis of housing expenditure in a self-selected
population of renters.
2. Binomial response models
2. I. Latent variable specification
The starting point for econometric analysis of a continuous response variable y is
often a linear regression model:
Y, = XrP- Et, ()
where x is a vector of exogenous variables, E is an unobserved disturbance, and
t=l , . . . , T indexes sample observations. The disturbances are usually assumed to
have a convenient cumulative distribution function F(E~x) such as multi-
variate normal. The model is then characterized by the conditional distribution
Ch. 24: Qualitat