*** CANCELLED *** Lefschetz Center for Dynamical Systems Seminar
Brown University Center for Statistical Sciences Seminar
Abstract: I have extended the general linear model methodology to a univariate generalized linear model based on permutation. This is a distribution-free approach to classical generalized linear models, as discussed by McCullagh and Nelder (1990), for nonlinear models where the expected response can be linearized with a link function. Two permutation schemes are examined along with the distributions of the test statistics produced by them, as discussed in papers by Kennedy (1995, 1996). I have applied this methodology to datasets which may help to clarify and enhance applicability of this methodology.
REFERENCES: McCullagh P. & Nelder J.A. (1990). Generalized Linear Models. Chapman & Hall: Cambridge.
Kennedy P.E. (1995). Randomization tests in econometrics. Journal of Business and Econometric Statistics, 13:85-94.
Kennedy P.E. & Cade, B.S. (1996). Randomizaton tests for multiple regression. Communications in Statistics, B25:923-936.
<--- 1999 Index