透過您的圖書館登入
IP:13.58.252.8
  • 期刊

Bayesian Analysis of Semiparametric Zero-Inflated Poisson Regression Models

半參數零膨脹泊松回歸模型貝葉斯分析

摘要


參數零膨脹泊松(ZIP)回歸模型已被廣泛用於分析包含許多零的計數數據。在參數ZIP回歸分析時,線性預測通常用於模擬協變量的影響,但是線性預測有時可能是不合適的。因此,為了對協變量的影響有好的描述效果,三次B樣條用於估計協變量的影響。一個貝葉斯估計方法被開發來替代傳統的最大似然方法去估計半參數模型參數。現實生活中的一組數據用於展示該方法的實用性。

並列摘要


Parametric zero-inflated Poisson (ZIP) regression models have been widely used to analyze count data that contains many zeros. In the parametric ZIP regression analysis, a linear predictor is usually used to model the effect of a covariate, which sometimes may be inappropriate. Thus cubic B-splines are used to approximate the covariate effect in order to describe the effect well. A Bayesian estimation method, as an alternative to traditional maximum likelihood-based methods, is developed to estimate the semiparametric model parameters. The practicality of the methodology is demonstrated with a real-life data set.

延伸閱讀