Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/192521 
Year of Publication: 
2008
Series/Report no.: 
Discussion Papers No. 539
Publisher: 
Statistics Norway, Research Department, Oslo
Abstract: 
Econometric duration data are typically interval-censored, that is, not directly observed, but observed to fall within a known interval. Known non-parametric identification results for duration models with unobserved heterogeneity rely crucially on exact observation of durations at a continuous scale. Here, it is established that the mixed hazards model is non-parametrically identified through covariates that vary over time within durations as well as between observations when durations are interval-censored. The results hold for the mixed proportional hazards model as a special case.
Subjects: 
duration analysis
interval-censoring
non-parametric identification
JEL: 
C41
Document Type: 
Working Paper

Files in This Item:
File
Size
195.32 kB





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.