Preprint #96-29
Estimation of P(Y>X) with Measurements Subject
to Error
by
Marie Coffin and Shashikala Sukhatme
Abstract
P(Y>X) can be estimated for data from any parametric family, although
unbiased estimates are often not available for data from lifetime
distributions, and the magnitude of bias in the estimate may be
considerable. Also, if the variables are measured with error, the
resulting estimates will be biased downward. Bias-corrected estimators
are found for normal and exponential models, and simulation studies
demonstrate that the bias-corrected estimators have MSE comparable
to that of the uncorrected estimators.
Copies of preprints are available from the author upon request. Use
the preprint number (located at the top of the page) and
make the request directly to the author, Iowa State
University,
Department of Statistics, Snedecor Hall, Ames, IA 50011-1210.