Richard V. Burkhauser, J. S. Butler, and Gulcin Gumus, "Dynamic
Programming Model Estimates of Social Security Disability Insurance
Application Timing," Journal of Applied Econometrics, Vol. 19, No. 6,
2004, pp. 671-685.

The data used in this paper ("HRS") are from "The Health and Retirement
Study: A Longitudinal Study of Health, Retirement, and Aging Sponsored by
the National Institute on Aging", available only from the Institute for
Social Research (ISR) of the University of Michigan.  Most of the data are
public use and are available from the website

  http://hrsonline.isr.umich.edu/
  
However, access to some of the data, namely the Social Security income
records and state of residence, is restricted and can be obtained under
certain conditions from the HRS staff at the Institute for Social Research
(ISR) at the University of Michigan.

See http://www.umich.edu/~hrswww/ for more information concerning
restricted data.

The data on initial acceptance rates in the Social Security Disability
Insurance program are available only from the Lewin Group.  See 

  Lewin Group (1995): Labor Economic Conditions, Socio-economic Factors, and
  the Growth of Applicants and Awards for SSDI and SSI Disability Benefits.
  Fairfax, VA: Lewin Group.  

Those data cannot be used without knowing the state of residence, which is
restricted access data in the HRS.

We provide a document, bbg-data-appendix.pdf, which explains the derivation
and estimation of the data, including estimated earnings, in detail. This
document, by Gulcin Gumus, is called "Data Appendix to Dynamic Programming
Model Estimates of Social Security Disability Insurance Application Timing".