Premium
Testing for Misspecification in a Censored‐Demand Model
Author(s) -
Tan Andrew,
Ozuna Teofilo
Publication year - 1994
Publication title -
applied economic perspectives and policy
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.4
H-Index - 49
eISSN - 2040-5804
pISSN - 2040-5790
DOI - 10.2307/1349470
Subject(s) - econometrics , economics
This article presents and applies conditional moment tests for detecting misspecification in a censored‐demand model. The seriousness of model misspecification is examined by comparing parameter estimates of a misspecified parametric model to those of a semiparametric model. Results indicate that the conditional moment tests are useful in detecting heteroscedasticity and non‐normality, and that misspecification can have an adverse effect on parameter estimates.