Alternative methods of estimating program effects in event history models

Curtis Eberwein, John Ham, Robert J. LaLonde

Research output: Contribution to journalArticle

Abstract

This paper first investigates the sensitivity of estimates of duration models to the specification of duration dependence. Using data from an experiment involving disadvantaged women in the U.S., we find that estimates of the parameters of hazard models are not sensitive to the way one models duration dependence as long as one uses a flexible functional form. We find that estimates of the expected duration in a state are insensitive to the way one models duration dependence if long spells are observed in the data, but that these are very sensitive to the specification when there are only relatively short spells in the data. We propose and implement alternative summary measures based on the median duration in a spell and show that these are quite robust to the specification of duration dependence.

Original languageEnglish (US)
Pages (from-to)249-278
Number of pages30
JournalLabour Economics
Volume9
Issue number2
DOIs
StatePublished - Jun 17 2002

Fingerprint

Duration dependence
Event history
Hazard models
Flexible functional forms
Duration models
Median
Experiment

Keywords

  • Duration models
  • Duration of dependence

ASJC Scopus subject areas

  • Economics and Econometrics
  • Organizational Behavior and Human Resource Management

Cite this

Alternative methods of estimating program effects in event history models. / Eberwein, Curtis; Ham, John; LaLonde, Robert J.

In: Labour Economics, Vol. 9, No. 2, 17.06.2002, p. 249-278.

Research output: Contribution to journalArticle

Eberwein, Curtis ; Ham, John ; LaLonde, Robert J. / Alternative methods of estimating program effects in event history models. In: Labour Economics. 2002 ; Vol. 9, No. 2. pp. 249-278.
@article{1114fdb888b44a01a77c5493959f1c76,
title = "Alternative methods of estimating program effects in event history models",
abstract = "This paper first investigates the sensitivity of estimates of duration models to the specification of duration dependence. Using data from an experiment involving disadvantaged women in the U.S., we find that estimates of the parameters of hazard models are not sensitive to the way one models duration dependence as long as one uses a flexible functional form. We find that estimates of the expected duration in a state are insensitive to the way one models duration dependence if long spells are observed in the data, but that these are very sensitive to the specification when there are only relatively short spells in the data. We propose and implement alternative summary measures based on the median duration in a spell and show that these are quite robust to the specification of duration dependence.",
keywords = "Duration models, Duration of dependence",
author = "Curtis Eberwein and John Ham and LaLonde, {Robert J.}",
year = "2002",
month = "6",
day = "17",
doi = "10.1016/S0927-5371(02)00005-2",
language = "English (US)",
volume = "9",
pages = "249--278",
journal = "Labour Economics",
issn = "0927-5371",
publisher = "Elsevier",
number = "2",

}

TY - JOUR

T1 - Alternative methods of estimating program effects in event history models

AU - Eberwein, Curtis

AU - Ham, John

AU - LaLonde, Robert J.

PY - 2002/6/17

Y1 - 2002/6/17

N2 - This paper first investigates the sensitivity of estimates of duration models to the specification of duration dependence. Using data from an experiment involving disadvantaged women in the U.S., we find that estimates of the parameters of hazard models are not sensitive to the way one models duration dependence as long as one uses a flexible functional form. We find that estimates of the expected duration in a state are insensitive to the way one models duration dependence if long spells are observed in the data, but that these are very sensitive to the specification when there are only relatively short spells in the data. We propose and implement alternative summary measures based on the median duration in a spell and show that these are quite robust to the specification of duration dependence.

AB - This paper first investigates the sensitivity of estimates of duration models to the specification of duration dependence. Using data from an experiment involving disadvantaged women in the U.S., we find that estimates of the parameters of hazard models are not sensitive to the way one models duration dependence as long as one uses a flexible functional form. We find that estimates of the expected duration in a state are insensitive to the way one models duration dependence if long spells are observed in the data, but that these are very sensitive to the specification when there are only relatively short spells in the data. We propose and implement alternative summary measures based on the median duration in a spell and show that these are quite robust to the specification of duration dependence.

KW - Duration models

KW - Duration of dependence

UR - http://www.scopus.com/inward/record.url?scp=0036267984&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036267984&partnerID=8YFLogxK

U2 - 10.1016/S0927-5371(02)00005-2

DO - 10.1016/S0927-5371(02)00005-2

M3 - Article

AN - SCOPUS:0036267984

VL - 9

SP - 249

EP - 278

JO - Labour Economics

JF - Labour Economics

SN - 0927-5371

IS - 2

ER -