Attribution Analysis of Changes in Climate Extremes

In inference for max-stable processes in regional frequency analysis, it is found that, when the dependence model is misspecified, the pairwise likelihood method leads to bias in estimating the shape parameter of the generalized extreme value (GEV) distribution. The bias can be serious when the dependence is strong. Motivated by the fact that the primary interest in many studies is the inference about marginal GEV parameters and that the spatial dependence is a nuisance, we propose a combined score equations (CSE) approach that does not need dependence assumptions beyond the univariate GEV distribution. The CSE method combines the score equations of GEV model at each site with an approximate correlation function of the scores to improve the estimation efficiency. Applied to fingerprinting of changes in climate extremes with a coordinate descent algorithm to estimate a large number of parameters, the CSE method provides a close analog to the optimal fingerprinting in detection and attribution of changes in Zhuo Wang – University of Connecticut, 2015 climate extremes. The approach is applied on extreme temperature in Australia unde