Bootstrapping the Conditional Moment Test for Parametric Duration Models

This letter evaluates the performance of auxiliary regression-based specification tests for parametric duration models estimated with censored data. The test using asymptotic critical values has poor size. Bootstrapping corrects the size problem but results in a biased power curve.

MoreLess

Year of publication:	2003-07-07
Authors:	Prieger, James E.
Institutions:	Economics Department, University of California-Davis
Subject:	conditional moment test \| test size \| right censoring \| type I censoring \| duration analysis \| exponential distribution \| Weibull distribution \| specification test \| power curve \| bootstrap bias

More details

Extent:	application/pdf
Series:	Working Papers.
Type of publication:	Book / Working Paper
Notes:	Number 34 9 pages long
Classification:	C12 - Hypothesis Testing ; C41 - Duration Analysis ; C24 - Truncated and Censored Models
Source:	RePEc - Research Papers in Economics

Persistent link: https://www.econbiz.de/10008620500