Journal ArticleDOI
Specification Error in Probit Models
Adonis Yatchew,Zvi Griliches +1 more
Reads0
Chats0
About:
This article is published in The Review of Economics and Statistics.The article was published on 1985-02-01. It has received 416 citations till now. The article focuses on the topics: Multinomial probit & Multivariate probit model.read more
Citations
More filters
Journal ArticleDOI
Back to the Future: Modeling Time Dependence in Binary Data
TL;DR: Monte Carlo analysis demonstrates that, for the types of hazards one often sees in substantive research, the polynomial approximation always outperforms time dummies and generally performs as well as splines or even more flexible autosmoothing procedures.
Book ChapterDOI
Measurement Error in Survey Data
TL;DR: While standard methods will not eliminate the bias when measurement errors are not classical, one can often use them to obtain bounds on this bias, and it is argued that validation studies allow us to assess the magnitude of measurement errors in survey data, and the validity of the classical assumption.
Journal ArticleDOI
Understanding heterogeneous preferences in random utility models: a latent class approach.
TL;DR: In this paper, a finite mixture approach to conditional logit models is developed in whichlatent classes are used to promoteunderstanding of systematic heterogeneity in wilderness recreation, and a branded choice experiment involvingchoice of one park from a demand system was administered to a sample of recreationists.
Journal ArticleDOI
Comparing Regression Coefficients Between Same-sample Nested Models Using Logit and Probit A New Method
TL;DR: The authors used logit and probit models to compare the coefficients of a given variable across different specified models fitted to the same set of variables. But the common practice of comparing the coefficients across different models is to compare a variable across a set of models.
Journal ArticleDOI
Measuring living standards with proxy variables
TL;DR: It is found that the proxies employed in much demographic research are very weak predictors of consumption per adult, Nevertheless, hypothesis tests based on proxies are likely to be powerful enough to warrant consideration.