利用右型II設限樣本對浴缸型分配的未來有序觀測值做貝氏預測區間估計

機構典藏 > College of Business and Management > Graduate Institute & Department of Statistics > Thesis > Item 987654321/74359

Please use this identifier to cite or link to this item: https://tkuir.lib.tku.edu.tw/dspace/handle/987654321/74359

Title:	利用右型II設限樣本對浴缸型分配的未來有序觀測值做貝氏預測區間估計
Other Titles:	Bayesian predictive interval estimation of future ordered observations for the bathtub-shaped distribution based on type II right censored samples
Authors:	王建智;Wang, Chien-Chih
Contributors:	淡江大學統計學系碩士班吳錦全;Wu, Chin-Chuan
Keywords:	浴缸型分配;右型II設限樣本;貝氏預測;蒙地卡羅模擬;Bathtub-shape distribution;Right Type II censored sample;Monte Carlo Simulation;Bayesian predictive
Date:	2011
Issue Date:	2011-12-28 18:24:43 (UTC+8)
Abstract:	近年來隨著科技的進步及時代的變遷，消費者對產品的使用期限及產品壽命有著更高的要求，希望能對產品壽命有著更多的保障。為了有效提升產品品質可靠度，常常需要做產品的抽樣壽命實驗，在壽命實驗中常因時間限制及人力和成本考量而無法取得完整的樣本資料，此時可以使用設限樣本。本文的主要目的是利用來自浴缸型分配的右型II設限樣本，並以貝式論點求取未來樣本觀測值的預測區間。內容包含對右型II設限樣本的假設、事前分配選取，事後分配的推導及求未來樣本觀測值之預測區間。並利用蒙地卡羅之模擬法來看，在95%和99%的信心水準下，以平均覆蓋真值的機率大小及平均的均方誤，來評估模擬結果的成效。經由模擬的結果顯示，在短期的預測之下可以達到不錯的預測結果，但在長期的情形下較為不穩定。 In recent years, with advances in technology and changing times, consumers use the product and the product life has a period of higher demand. In order to effectively improve product quality and reliability. In lifetime testing experiment, we may not be able to obtain a complete sample due to time limitation or other restriction. Therefore, censored samples arise in practices. This research use Bayesian predictive interval estimation of future ordered observations for the bathtub-shaped distribution based on Type II right censored samples. And use Monte Carlo simulation , in 95% and 99% confidence level, the average coverage probability and the mean square error to evaluate the effectiveness of simulation results. The results show that, under the short-term forecasts predict a good outcome can be achieved, but in the case of a more long-term instability.
Appears in Collections:	[Graduate Institute & Department of Statistics] Thesis

Files in This Item:

File	Size	Format
index.html	0Kb	HTML	306	View/Open

Loading...