Methodological research in statistics is motivated often by interest in methodological/theoretical statistical questions. Consulting and collaborative projects also motivate the research because existing methods and theories may be inadequate to handle their data analysis or study design. Recently Dr. Lee completed a paper on approximating P-values after repeated significance testing. Methodological research in missing data in clinical trials and also in estimating confidence intervals for the difference in survival rates of two populations is ongoing. Such research is motivated more by general methodological interest in statistics and biostatistics because resulting methods are more broadly applicable. Dr. Park has been actively engaged in regression models for repeated measurements with or without missing data. Dr. Troendle is working on time series methods. Dr. Troendle is studying the statistical power of an existing method when it is applied to the data with many (emphasis added) outcome variables. This research is motivated by the need to analyze the DIEP study where 55 types of minor malformations were studied to determine the effect of mothers' diabetes on her baby. Drs. Lee and Park have been working on statistical methods of analyzing time to extubation in premature babies with intubation for mechanical ventilation while some babies die before the extubation. These are some of the problems encountered in the course of collaboration and consulting. Innovative modification and application as well as new methods are required to handle such data.
