Statistics Seminars: Goodness-of-fit tests based on the empirical characteristic function: A review
14 November 2008 14:00 in CM221
Testing procedures are reviewed that incorporated the characteristic function and its empirical counterpart. The methods are seen as counterparts in the Fourier domain of classical procedures based on the empirical distribution function. We start with classical goodfness-of-fit problems for parametric families of distributions, and continue with the same problem i) with nuisance regression parameters, ii) in the non-parametric regression case and iii) in the semi-parametric context. We finally cite some other problems in which the Fourier approach has been found application, such as testing for symmetry and independence, as well as the two-sample and the change-point problem. The presentation focuses on basic principles underlying the procedures rather than being technical.
Contact firstname.lastname@example.org for more information