The Econometric Analysis of Transition Data

This is the first published exposition of current econometric methods for the study of duration data.

Tony Lancaster (Author)

9780521437899, Cambridge University Press

Paperback, published 26 June 1992

368 pages, 23 b/w illus. 1 table
22.6 x 15.1 x 1.8 cm, 0.54 kg

'The analysis of transition/duration data is the next frontier beyond the now well-charted area of qualitative choice and limited-dependent variables. Lancaster's book is an excellent travel guide to this relatively unknown but beautiful land where time-series analysis and qualitative choice merge. Even if they do not decide to stay, economists and social scientists will return enriched from a trip under Professor Lancaster's guidance.' Geert Ridder, University of Groningen

This book presents statistical methods for analysis of the duration of events. The primary focus is on models for single-spell data, events in which individual agents are observed for a single duration. Some attention is also given to multiple-spell data. The first part of the book covers model specification, including both structural and reduced form models and models with and without neglected heterogeneity. The book next deals with likelihood based inference about such models, with sections on full and semiparametric specification. A final section treats graphical and numerical methods of specification testing. This is the first published exposition of current econometric methods for the study of duration data.

Preface
Part I. Model Building: 1. Some basic results
2. Covariates and the hazard function
3. Parametric families of duration distribution
4. Mixture models
5. Some important processes
6. Some structural transition models
Part II. Inference: 7. Identifiability issues
8. Fully parametric inference
9. Limited information inference
10. Misspecification analysis
11. Residual analysis
Appendix 1: The gamma function and distribution
Appendix 2: Some properties of the Laplace transform
Bibliography
Index.

Subject Areas: Econometrics [KCH]