Michele Campolieti, "Bayesian Semiparametric Estimation of Discrete Duration Models: An Application of the Dirichlet Process Prior Model", Journal of Applied Econometrics, Vol. 16, No. 1, 2001, pp. 1-22. The data were obtained from the master files of the Labour Market Activity Survey (LMAS) and originally used by Michael Baker and Samuel A. Rea Jr., (1998), 'Employment Spells and Unemployment Insurance Eligibility Requirements,' Review of Economics and Statistics 80, 80-94. The data contract for access to the master files was signed by Samuel A. Rea Jr., Department of Economics, University of Toronto. Please direct inquiries regarding access to the LMAS master files to: Labour and Household Survey's Branch Statistics Canada Ottawa, Ontario, Canada K1A OT6 Tel: (613)-951-0053 The data file is formatted as follows: 5f1.0,F2.0,F4.0,F6.0,F8.0,F4.1,F8.2,13F1.0,F10.4,F1.0,2F3.0 and contains the following variables binary indicator - equals 1 if continue, 0 if exit EL1 EL2 EL3 Year Dummy Time Time^2 Time^3 Time^4 Unemployment rate Hourly earnings Age 16-24 Age 25-44 High School Post Secondary University Trade Certificate Past UI Receipt Marital Status School Attendance Gender IL1 - see Baker and Rea (1998) IL2 - see Baker and Rea (1998) IL3 - see Baker and Rea (1998) Individual Indicator (Individual Number) Variable definitions are provided in Table 1 of my paper, unless noted otherwise. Please address any further questions to: Michele Campolieti Centre for Industrial Relations University of Toronto 121 St. George Street M5S 2E8 Phone: (416)-978-2748 Fax: (416)-978-5696 Email: campolie@chass.utoronto.ca