This project addresses statistical problems generated from collaboration with scientists in other program areas and general statistical problems of current interest. This project is a continuing activity of the Section on Mathematical Statistics. Papers have been submitted or published in FY '86 on the following statistical subjects: pooled adjacent violators; general test of trend for count data; hypothesis tests involving linear inequality constraints; sampling strategies for rare diseases; the analysis of Phase II clinical trials; the development of robust selection procedures based on vector ranks; the evaluation of sample size and power for analysis of survival with allowance for non-uniform patient entry, losses to follow-up, non-compliance and stratification; and stochastic curtailing for comparison of slopes in longitudinal studies. Other work in progress includes: statistics for the evaluation of space-time clustering of disease; modeling of residual treatment effect for k-period two-treatment crossover designs; adjustments for covariates in the analysis of categorical data; the influence of missing data on statistical models determined by variable selection procedures; the effects of misclassification of exposure variables on case-control studies; the use of area surveys in epidemiological research; the impact of risk and incidence of risk on selection of variables in multiple regression; the development of methods to improve coverage in surveys; the use of non-parametric tests for umbrella type alternative hypotheses; the impact of informative censoring on survival analysis with time dependent covariates; and the development of methods of survival analysis for dependent competing risks.
