This project addresses statistical problems generated from collaboration with scientists in other program areas and general statistical problems of current interest. his project is a continuing activity of the Section on Mathematical Statistics and other members of the Branch. In FY 1995, Branch statisticians have contributed to the following areas of statistical research: evaluation of proxy respondents and auxiliary information as adjustments for nonresponse or attrition in disease surveys ; modeling seasonal change in time series regression relationships; development of both discrete and continuous time Markov models for longitudinal categorical data which accounts for different processes across subjects; derivation of statistical methods for the detection of and tests of differences among spatial disease clusters; estimation of selection bias and its effect on inferences from observational data bases; development of adjustment methods to account for differential mortality for the evaluation of factors associated with increased disease risk. Other work includes: methods to improve coverage in surveys; estimation of time-to-event data with interval censoring; site selection for epidemiologic surveys; analysis of response surface data with spatial and temporal components; modeling of response surfaces with spatially correlated errors; application of splines to estimate model parameters of multiple correlated response surfaces; modeling effect changes of covariates in the presence of spatial correlation; analysis of bioequivalence trials with multiple, nonlinear responses to treatment; combining information from negatively correlated nonlinear regressions; development of a generalized estimating equation approach for the analysis of spatially dependent binary data; application of bootstrap methods to longitudinal natural history data; use of variance component methods to assess the precision of biochemical measurements; using a Markov chain model to study three state disease processes; evaluation of case ascertainment strategies for area surveys of neurological diseases; and sampling strategies for spatial point processes with multiple types of clustering.
