Paul Devereux, "Small Sample Bias in Synthetic Cohort Models of Labor
Supply", Journal of Applied Econometrics, Vol. 22, No. 4, 2007, pp. 
839-848.

The dataset is an extract from the NBER CPS MORG files using years from
1979-1993. There are 498,667 observations and six variables in the file
pd-data.txt, which is an ASCII file in DOS format. This file is zipped in
the file pd-data.zip. Unix users should use "unzip -a".

The six variables, in order, are:

HOURS:	Weekly hours worked
LNWAGE:	Log of Hourly Wage
OTHINC:	Other Income
YEAR:	Survey Year
COHORT:	Birth Cohort
HSMORE:	Indicator Variable (1: More than High School; 0: High School or Less)