U.S.-Hungary Statistics Research: Topics in Change Point and Unit Root Analysis; Rates of Convergence, Permutations and Bootstrap

University Of Utah, Salt Lake City UT

Investigators

Abstract

This US-Hungarian collaborative project between Lajos Horvath and Piotr Kokoszka of the University of Utah and their Hungarian partners, Istvan Berkes and Endre Csaki of the A. Renyi Institute of Mathematics in Budapest, examines several statistical and probabilistic problems concerned with the detection and analysis of nonstationarities of various types of stochastic processes. Emphasis is on careful analysis of the rates of convergence of the statistics of interest and on the properties of the associated permutation and bootstrap procedures. Junior researchers from the University of Utah also participate. The joint research plan includes an investigation of change point procedures and the properties of unit root tests, that are to be developed based on resampling and subsampling methods. These are fundamental questions associated with the detection of time series. Results may have practical applications in economics and the financial sector by leading to useful derivations of properties of CUSUM procedures and GARCH models. For example, findings may improve models for stock returns that can be refined to accommodate infinite error variance. This statistics research project fulfills the program objective of advancing scientific knowledge by enabling experts in the United States and Central Europe to combine complementary talents and share research resources in areas of strong mutual interest and competence.