Regression trees for analysis of count data with extra Poisson variation

摘要: This article proposes methods for fitting piecewise loglinear models to count data with an extra-Poisson variation. Both SUPPORT (Statistica Sinica, 4 (1994) 143) and GUIDE 12 (2002) 361) are used splitting methods. We developed a new bootstrap resampling method performed at each node of the tree determine proper size tree. The quasi-likelihood approach is model stratum take into account extra variability. An adjusted Anscombe residual in this procedure. Performance proposed evaluated by Monte Carlo simulation study. investigate geographic variability mortality rates on lung cancer as well effects various demographic