The research project involves statistical resampling methods, analysis of directional data, survival analysis and semiparametric models. The development of a unified theory of resampling procedures that covers both the bootstrap and the jackknife methods will be supported. The resulting elements from k sampled blocks from a sequence of stationary m-dependent random variables will be used to provide bootstrap estimators of parameters and the properties of the estimators will be examined. Questions of robustness of these resampling procedures will be investigated. Notions of data depth for directional data will be developed along with nonparametric tests for them and corresponding classifications of data. Semiparametric models for survival analysis will be be studied for both censored and non- censored data.

Agency
National Science Foundation (NSF)
Institute
Division of Mathematical Sciences (DMS)
Application #
9004658
Program Officer
Alan Izenman
Project Start
Project End
Budget Start
1990-06-15
Budget End
1993-11-30
Support Year
Fiscal Year
1990
Total Cost
$158,235
Indirect Cost
Name
Rutgers University
Department
Type
DUNS #
City
New Brunswick
State
NJ
Country
United States
Zip Code
08901