Bootstrap Methods for Data with Multiple Errors

This project will develop and implement a range of sample resampling or bootstrap methods for the analysis of data whose distribution is affected by multiple sources of variation. Bootstrap methods are computational methods for making statistical inferences when samples are small and other approximations work poorly, two situations which often arise with data with multiple errors. Such data arise in all types of statistical studies carried out in diverse fields of application so the results of the project will have wide applicability. The expected outcomes are a both a deeper understanding based on theoretical results and practical tools for important problems expressed in theoretical papers and software respectively.