頁籤選單縮合
題 名 | 多重轉折點貝氏分析之應用=Applications for Multiple Changepoints Via Bayesian Analysis |
---|---|
作 者 | 劉淑鶯; 紀穎鴻; | 書刊名 | 中國統計學報 |
卷 期 | 37:2 1999.06[民88.06] |
頁 次 | 頁161-183 |
分類號 | 319.16 |
關鍵詞 | 轉折點; 卜瓦松過程; 後驗分配; 可逆跳躍式馬可夫鏈蒙地卡羅法; Changepoints; Poisson process; Posterior distribution; Veversible jump markov chain; Monte carlo method; |
語 文 | 中文(Chinese) |
中文摘要 | 本文主要是針對Green(1995)所提出的可逆跳躍式馬可夫鏈蒙地卡羅 (Reversible Jump Markov Chain Monte Carlo) 的方法,及其在具多重轉折點卜瓦松模型上的應用給予較詳細之整理與介紹。對摸擬資料及實際資料,估計轉折點個數之後驗分配。並探討轉折點發生的位置、事件發生率等參數之條件後驗分配的一些統計量。最後對於實際資料可利用當時之實際情況對於轉折點發生之原因做適當的闡釋。 |
英文摘要 | This paper provides detailed decsriptions for the reversible jump Markov Chain Monte Carlo method and its application in the Poisson process with multiple changepoints, which was proposed by Green (1995). Both simulation data and real data are used to estimate the posteroir distribution of the number of changepoints. Moverover, some statistics of the conditional posterior distributions of the position of changepoints, and the occurrence rates of event are investigated. Finally, the interpretations of the occurrence of changepoints for the real data are appropriately explained. |
本系統中英文摘要資訊取自各篇刊載內容。