Empirical Economics

, Volume 39, Issue 1, pp 111–137 | Cite as

Identification of the effects of dynamic treatments by sequential conditional independence assumptions

Original Paper

Abstract

This paper approaches the dynamic analysis of the effects of training programs for the unemployed in West Germany, or in general the effects of sequences of interventions, from a potential outcome perspective. The identifying power of different assumptions concerning the connection between the dynamic selection process and the outcomes of different sequences is discussed. When participation in the particular sequence of programs is decided period by period depending on its success so far, many parameters of interest are no longer identified. Nevertheless, some interesting dynamic forms of the average treatment effect are identified by a sequential randomization assumption. Based on this approach, we present some new results on the effectiveness of West-German training programs.

Keywords

Labor market effects of training programs Dynamic treatment regimes Nonparametric identification Causal effects Sequential randomization Program evaluation Treatment effects Dynamic matching Panel data 

JEL Classification

C21 C31 J68 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Abbring JH (2003) Dynamic econometric program evaluation. IZA, Discussion Paper 804Google Scholar
  2. Abbring JH, Heckman JJ (2008) Dynamic policy analysis. In: Matyas L, Sevestre P (eds) The econometrics of panel data, 3rd edn. Kluwer, DordrechtGoogle Scholar
  3. Angrist JD (1998) Estimating labor market impact of voluntary military service using social security data. Econometrica 66: 249–288CrossRefGoogle Scholar
  4. Angrist JD, Imbens GW, Rubin DB (1996) Identification of causal effects using instrumental variables. J Am Stat Assoc 91: 444–472CrossRefGoogle Scholar
  5. Behrman J, Cheng Y, Todd P (2004) Evaluating preschool programs when length of exposure to the program varies: a nonparametric approach. Rev Econ Stat 86: 108–132CrossRefGoogle Scholar
  6. Behrman J, Sengupta P, Todd P (2005) Progressing through PROGRESA: an impact assessment of a school subsidy experiment in Mexico. Econ Dev Cult Change 54/1: 237–275CrossRefGoogle Scholar
  7. Bergemann A, Fitzenberger B, Speckesser S (2004) Evaluating the dynamic employment effects of training programs in East Germany using conditional difference-in-differences. ZEW, Discussion Paper 04-41Google Scholar
  8. Brodaty T, Crépon B, Fougère D (2001) Using matching estimators to evaluate alternative youth employment programmes: evidence from France, 1986–1988. In: Lechner M, Pfeiffer F (eds) Econometric Evaluation of Labour Market Policies. Physica, Heidelberg, pp 85–123Google Scholar
  9. Cameron CA, Trivedi PK (2005) Microeconometrics, Chap. 25, CUP, CambridgeGoogle Scholar
  10. Chamberlain G (1987) Asymptotic efficiency in estimation with conditional moment restrictions. J Econom 34: 305–334CrossRefGoogle Scholar
  11. Chamberlain G (1992) Comment: sequential moment restrictions in panel data. J Bus Econ Stat 10: 20–26CrossRefGoogle Scholar
  12. Cochran WG, Chambers SP (1965) The planning of observational studies of human population. J R Stat Soc Ser A 128: 234–266CrossRefGoogle Scholar
  13. Crépon B, Kramarz F (2002) Employed 40 hours or not-employed 39: lessons from the 1982 workweek reduction in France. J Polit Econ 110:1355–1389Google Scholar
  14. Dawid AP (1979) Conditional independence in statistical theory. J R Stat Soc Ser B 41: 1–31Google Scholar
  15. Ding W, Lehrer SF (2003) Estimating dynamic treatment effects from project STAR. MimeoGoogle Scholar
  16. Gerfin M, Lechner M (2002) Microeconometric evaluation of the active labour market policy in Switzerland. Econ J 112: 854–893CrossRefGoogle Scholar
  17. Gill RD, Robins JM (2001) Causal inference for complex longitudinal data: the continuous case. Ann Stat 29: 1785–1811CrossRefGoogle Scholar
  18. Heckman JJ, Vytlacil E (1999) Local instrumental variables and latent variable models for identifying and bounding treatment effects. Proc Natl Acad Sci 96:4730–4734Google Scholar
  19. Heckman JJ, Vytlacil E (2006) Econometric evaluation of social programs. Heckman JJ, Leamer E (eds) Handbook of Econometrics, vol. VI. North-Holland, Amsterdam (forthcoming)Google Scholar
  20. Heckman JJ, Robb R (1985) Alternative methods of evaluating the impact of interventions. In: Heckman JJ, Singer B (eds) Longitudinal analysis of labour market data. Cambridge University Press, New York, pp 156–245CrossRefGoogle Scholar
  21. Heckman JJ, Navarro S (2005) Dynamic discrete choice and dynamic treatment effects. IZA discussion paper 1790Google Scholar
  22. Heckman JJ, LaLonde RJ, Smith JA (1999) The economics and econometrics of active labor market programs. In: Ashenfelterc O, Card D (eds) Handbook of labor economics, vol III A. North-Holland, Amsterdam, pp 1865–2097Google Scholar
  23. Imbens GW (2000) The role of the propensity score in estimating dose-response functions. Biometrika 87: 706–710CrossRefGoogle Scholar
  24. Imbens GW (2004) Nonparametric estimation of average treatment effects under exogeneity: a review. Rev Econ Stat 86: 4–29CrossRefGoogle Scholar
  25. Imbens GW, Angrist JD (1994) Identification and estimation of local average treatment effects. Econometrica 62: 446–475CrossRefGoogle Scholar
  26. Lechner M (1999) Earnings and employment effects of continuous off-the-job training in East Germany after unification. J Bus Econ Stat 17: 74–90CrossRefGoogle Scholar
  27. Lechner M (2001) Identification and estimation of causal effects of multiple treatments under the conditional independence assumption. In: Lechner M, Pfeiffer F (eds) Econometric evaluation of active labour market policies. Physica, Heidelberg, pp 43–58Google Scholar
  28. Lechner M (2002) Programme heterogeneity and propensity score matching: an application to the evaluation of active labour market policies. Rev Econ Stat 84: 205–220CrossRefGoogle Scholar
  29. Lechner M (2004) Sequential matching estimation of dynamic causal models. University of St. Gallen, Discussion paper, 2004-06; Revised and shortened version. J Bus Econ Stat (forthcoming)Google Scholar
  30. Lechner M (2008a) A note on endogenous control variables in evaluation studies. Stat Probab Lett 78: 190–195CrossRefGoogle Scholar
  31. Lechner M (2008b) Matching estimation of dynamic treatment models: some practical issues. In: Millimet D, Smith J, Vytlacil E (eds) Advances in econometrics, vol 21. Modelling and evaluating treatment effects in econometrics, pp 289–333Google Scholar
  32. Lechner M, Miquel R (2001) A potential outcome approach to dynamic programme evaluation: non-parametric identification, Department of Economics, University of St. Gallen, 2001-07Google Scholar
  33. Lechner M, Miquel R, Wunsch C (2005) Long—run effects of public sector—sponsor training in West Germany. IAB, Discussion Paper 3/2005Google Scholar
  34. Manski CF (2004) Social learning from private experiences: the dynamics of the selection problem. J Econ Stud 71: 443–558CrossRefGoogle Scholar
  35. Miquel R (2002) Identification of dynamic treatments effects by instrumental variables. University of St. Gallen, Discussion paper 2002-11Google Scholar
  36. Miquel R (2003) Identification of effects of dynamic treatments with a difference-in-differences approach. University of St. Gallen, Discussion paper 2003-06Google Scholar
  37. Murphy SA (2003) Optimal dynamic treatment regimes. J R Stat Soc Ser B 65: 331–366CrossRefGoogle Scholar
  38. Neyman J (1923) On the application of probability theory to agricultural experiments. Essay on principles. Section 9, translated in Statistical Science (with discussion) (1990) 5:465–480Google Scholar
  39. Robins JM (1999) Association, causation, and marginal structural models. Synthese 121: 151–179CrossRefGoogle Scholar
  40. Robins JM (1986) A new approach to causal inference in mortality studies with sustained exposure periods—application to control of the healthy worker survivor effect. Math Model 7:1393–1512, with 1987 Errata to “A new approach to causal inference in mortality studies with sustained exposure periods—application to control of the healthy worker survivor effect”. Comput Math Appl 14:917–921; 1987 Addendum to “A new approach to causal inference in mortality studies with sustained exposure periods—application to control of the healthy worker survivor effect”, Comput Math Appl 14:923–945; and 1987 Errata to “Addendum to ‘A new approach to causal inference in mortality studies with sustained exposure periods—application to control of the healthy worker survivor effect”’ Comput Math Appl 18:477Google Scholar
  41. Robins JM (1989) The analysis of randomized and nonrandomized aids treatment trials using a new approach to causal inference in longitudinal studies. In: Sechrest L, Freeman H, Mulley A (eds) Health service research methodology: a focus on AIDS, 113–159, Washington, D.C., Public Health Service, National Center for Health Services ResearchGoogle Scholar
  42. Robins JM (1997) Causal inference from complex longitudinal data. Latent variable modelling and applications to causality. In: Berkane M (ed) Lecture Notes in Statistics, vol 120. Springer, New York, pp 69–117Google Scholar
  43. Robins JM (1998) Marginal structural models. In: 1997 Proceedings of the American Statistical Association, pp 1–10Google Scholar
  44. Robins JM, Hernan MA, Brumback B (2000) Marginal structural models and causal inference in epidemiology. Epidemiology 11(5): 550–560CrossRefGoogle Scholar
  45. Robins JM, Greenland S, Hu F (1999a) Estimation of the causal effect of a time-varying exposure on the marginal mean of a repeated binary outcome. J Am Stat Assoc 94: 687–700CrossRefGoogle Scholar
  46. Robins JM, Greenland S, Hu F (1999b) Estimation of the causal effect of a time-varying exposure on the marginal mean of a repeated binary outcome: rejoinder. J Am Stat Assoc 94: 708–712CrossRefGoogle Scholar
  47. Rosenbaum PR (1984) The consequences of adjustment for a concomitant variable that has been affected by the treatment. J R Stat Soc Ser A 147: 656–666CrossRefGoogle Scholar
  48. Rosenbaum PR, Rubin DB (1983) The central role of the propensity score in observational studies for causal effects. Biometrika 70: 41–50CrossRefGoogle Scholar
  49. Roy AD (1951) Some thoughts on the distribution of earnings. Oxf Econ Pap 3: 135–146Google Scholar
  50. Rubin DB (1974) Estimating causal effects of treatments in randomized and nonrandomized studies. J Educ Psychol 66: 688–701CrossRefGoogle Scholar
  51. Rubin DB (1977) Assignment to a treatment group on the basis of a covariate. J Educ Stat 2: 1–26CrossRefGoogle Scholar
  52. Rubin DB (2004) Direct and indirect causal effects via potential outcomes. Scand J Stat 31: 161–170CrossRefGoogle Scholar
  53. Rubin DB (2005) Causal inference using potential outcomes: design, modeling, decisions. J Am Stat Soc 100: 322–331Google Scholar
  54. Sianesi B (2004) An evaluation of the Swedish System of active labour market programmes in the 1990s. Rev Econ Stat 86: 133–155CrossRefGoogle Scholar
  55. Taber CR (2000) Semiparametric identification and heterogeneity in discrete choice dynamic programming models. J Econom 96: 201–229CrossRefGoogle Scholar
  56. Vytlacil E (2002) Independence, monotonicity, and latent index models: an equivalence result. Econometrica 70: 331–341CrossRefGoogle Scholar
  57. Wilks SS (1932) On the distribution of statistics in samples from a normal population of two variables with matched sampling of one variable. Metron 9: 87–126Google Scholar

Copyright information

© Springer-Verlag 2009

Authors and Affiliations

  1. 1.Swiss Institute for Empirical Economic Research (SEW)University of St. GallenSt. GallenSwitzerland

Personalised recommendations