2009년 11회 통계세미나 개최 안내
통계연구소에서는 다음과 같이 통계 세미나를 개최하오니 많은 참여 바랍니다.
일시: 2009년 12월 9일(수) 오후 5시
장소: 고려대학교 정경관 507호
연사: Hyun-Joo Kim
(Department of Mathematics and Computer Science,
Truman State University)
“Model Selection Criteria for a Counting Data with
Overdispersion and its Application to the Host-Parasite
Relationship”
In statistical modeling, selecting an optimal model from a class of candidates is a critical issue. During the past four decades, a number of model selection criteria have been widely used based on estimating Kullback's (1968, p. 5) directed or symmetric divergence including the Akaike (1973, 1974) information criterion, AIC, the ``corrected" Akaike information criterion (Hurvich and Tsai, 1989), AICc, Kullback information criterion, KIC, and ``corected" Kullback information criterion, KICc (Cavanaugh, 1999, 2003). In wildlife and ecology, a counting data dealing with overdispersion is common. For such case, a simple modification of AIC and AICc, Quasi Akaike information criteria and its corrected version, QAIC and QAICc (1992) have been introduced. In this paper, new criteria, QKIC and QKICc are proposed based on Kullback symmetric divergence as analogues of QAIC and QAICc. The selection performance of AIC, AICc, QAIC, QAICc, KIC, KICc, QKIC, and QKICc is evaluated in a simulation study. It is also applied to develop a statistical model of the tick, Dermacentor variabilis,load on a white-footed mouse, Peromyscus leucopus in northern Missouri.
고려대학교 통계연구소