南阳师范学院学报
南暘師範學院學報
남양사범학원학보
JOURNAL OF NANYANG TEACHERS COLLEGE
2011年
12期
18-22
,共5页
零膨胀%负二项分布%回归模型%索赔频率
零膨脹%負二項分佈%迴歸模型%索賠頻率
령팽창%부이항분포%회귀모형%색배빈솔
zero-inflated%negative binomial%regression model%claim frequency
讨论了两种分布形式的零膨胀负二项回归模型，并应用一组实际汽车保险损失数据对两类模型进行了实证比较．结果表明，对于具有零膨胀特征的损失数据，零膨胀负二项回归模型的拟合结果优于普通索赔频率回归模型．
討論瞭兩種分佈形式的零膨脹負二項迴歸模型，併應用一組實際汽車保險損失數據對兩類模型進行瞭實證比較．結果錶明，對于具有零膨脹特徵的損失數據，零膨脹負二項迴歸模型的擬閤結果優于普通索賠頻率迴歸模型．
토론료량충분포형식적령팽창부이항회귀모형，병응용일조실제기차보험손실수거대량류모형진행료실증비교．결과표명，대우구유령팽창특정적손실수거，령팽창부이항회귀모형적의합결과우우보통색배빈솔회귀모형．
The paper discusses two types of distribution forms zero-inflated negative binomial regression models, and applies the two models to a set of actual automobile insurance loss data. The results show that the two types of zero-inflated negative binomial regression models can improve the goodness-of-fit effectively than the common claim frequency models when the loss data is zero-inflated.
