The nonparametric identification of treatment effects in duration models

This paper analyzes the specification and identification of causal multivariate duration models. We focus on the case in which one duration concerns the point in time a treatment is initiated and we are interested in the effect of this treatment on some outcome duration. We define "no anticipation of treatment" and relate it to a common assumption in biostatistics. We show that (i) no anticipation and (ii) randomized treatment assignment can be imposed without restricting the observational data. We impose (i) but not (ii) and prove identification of models that impose some structure. We allow for dependent unobserved heterogeneity and we do not exploit exclusion restrictions on covariates. We provide results for both single-spell and multiple-spell data. The timing of events conveys useful information on the treatment effect. Copyright The Econometric Society 2003.

[1]  Tony Lancaster,et al.  The Econometric Analysis of Transition Data. , 1992 .

[2]  D. Harrington,et al.  Counting Processes and Survival Analysis , 1991 .

[3]  Denis Fougère,et al.  Evaluating the Impact of French Employment Policies on Individual Labour Market Histories , 1997 .

[4]  L A Lillard,et al.  Simultaneous equations for hazards: marriage duration and fertility timing. , 1993, Journal of econometrics.

[5]  D. Rubin.  [On the Application of Probability Theory to Agricultural Experiments. Essay on Principles. Section 9.] Comment: Neyman (1923) and Causal Inference in Experiments and Observational Studies , 1990 .

[6]  Jaap H. Abbring,et al.  A simple procedure for inference on treatment effects in duration models , 2004 .

[7]  C. Manski.  Nonparametric Bounds on Treatment Effects , 1989 .

[8]  J. Heckman,et al.  The Economics and Econometrics of Active Labor Market Programs , 1999 .

[9]  James J. Heckman,et al.  The identifiability of the competing risks model , 1989 .

[10]  D. Cox,et al.  THE ANALYSIS OF EXPONENTIALLY DISTRIBUTED LIFE-TIMES WITH Two TYPES OF FAILURE , 1959 .

[11]  Geert Ridder,et al.  The Non-Parametric Identification of Generalized Accelerated Failure-Time Models , 1990 .

[12]  C. Glymour,et al.  STATISTICS AND CAUSAL INFERENCE , 1985 .

[13]  G. Ridder,et al.  Stratified partial likelihood estimation. , 1999, Journal of econometrics.

[14]  Gerard J. van den Berg,et al.  The identifiability of the mixed proportional hazards competing risks model , 2003 .

[15]  Gerard J. van den Berg,et al.  Erratum to: Analyzing the effect of dynamically assigned treatments using duration models, binary treatment models, and panel data models , 2018 .

[16]  James M. Robins,et al.  Causal inference for complex longitudinal data: the continuous case , 2001 .

[17]  R. M. Gritz,et al.  The impact of training on the frequency and duration of employment , 1993 .

[18]  Rupert G. Miller,et al.  Survival Analysis , 2022, The SAGE Encyclopedia of Research Design.

[19]  C. Manski.  Monotone Treatment Response , 2009, Identification for Prediction and Decision.

[20]  David R. Cox,et al.  Regression models and life tables (with discussion , 1972 .

[21]  Joel L. Horowitz,et al.  Semiparametric Estimation of a Proportional Hazard Model with Unobserved Heterogeneity , 1999 .

[22]  Patrick Billingsley,et al.  Probability and Measure. , 1986 .

[23]  Jan C. van Ours,et al.  The Effect of Unemployment Insurance Sanctions on the Transition Rate from Unemployment to Employment , 1996 .

[24]  David Card,et al.  Measuring the Effect of Subsidized Training Programs on Movements in Andout of Employment , 1987 .

[25]  Dale T. Mortensen,et al.  Chapter 15 Job search and labor market analysis , 1986 .

[26]  Douglas R. Miller.  A Note on Independence of Multivariate Lifetimes in Competing Risks Models , 1977 .

[27]  P. Sorlie,et al.  Marital status and mortality: the national longitudinal mortality study. , 2000, Annals of epidemiology.

[28]  D. Cox.  Regression Models and Life-Tables , 1972 .

[29]  J. Pearl.  Causality: Models, Reasoning and Inference , 2000 .

[30]  Chris Elbers,et al.  True and Spurious Duration Dependence: The Identifiability of the Proportional Hazard Model , 1982 .

[31]  J. Kalbfleisch,et al.  The Statistical Analysis of Failure Time Data , 1980 .

[32]  Gerard J. van den Berg,et al.  Punitive sanctions and the transition rate from welfare to work , 1998 .

[33]  D. Rubin.  Estimating causal effects of treatments in randomized and nonrandomized studies. , 1974 .

[34]  Stayers versus defecting movers: a note on the identification of defective duration models , 2002 .

[35]  P. Lichtenstein,et al.  A twin study of mortality after spousal bereavement , 1998, Psychological Medicine.

[36]  J. D. Holt,et al.  Survival analyses in twin studies and matched pair experiments , 1974 .

[37]  J J Heckman,et al.  Econometric mixture models and more general models for unobservables in duration analysis , 1994, Statistical methods in medical research.

[38]  Gerard J. van den Berg,et al.  Duration models: specification, identification and multiple durations , 2000 .

[39]  Bo E. Honoré,et al.  IDENTIFICATION RESULTS FOR DURATION MODELS WITH MULTIPLE SPELLS OR TIME-VARYING COVARIATES , 1990 .

[40]  A. Dawid.  Causal Inference without Counterfactuals , 2000 .