Title: A generalized linear model for extreme events: New goodness of fit measures
Authors: Francesco Zuniga - University of Nevada at Reno (United States) [presenting]
Anna Panorska - University of Nevada (United States)
Tomasz Kozubowski - University of Nevada Reno (United States)
Abstract: A generalized linear model is presented for the observed vector $(N, X, Y)$ of duration $N$, magnitude $X$ and maximum $Y$ of an (often extreme) event such as flood, extreme precipitation, market growth or decline. Out GLM allows for the investigation of the association of the $(N, X, Y)$ with covariates. We also present new approach of goodness of fit assessment. The methodology is illustrated by modeling growth events of the S\&P500 index on covariates such as unemployment and inflation rate.