Please use this identifier to cite or link to this item:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/1203
Title: | 建立在伽瑪散度下穩健模型的調整參數之選取 Robust Model Fitting - Selection of Tuning Parameters in the Aspect of Gamma Clustering |
Authors: | Yi Hsiao 蕭奕 |
Advisor: | 杜憶萍(I-Ping Tu) |
Keyword: | 穩健估計,影響函數,伽瑪散度,群聚分析, robust estimate,influence function,gamma-divergence,clustering, |
Publication Year : | 2018 |
Degree: | 碩士 |
Abstract: | Windham在1995年的論文Robustifying Model Fitting提出權重分布方法,解決在存有異常數據情況下可有效的估計均值,這個權重分布使用了一個參數,這個參數會影響到均值的估計表現。Windham同時提出此參數的選取方法,但我們發現在一些數值模擬例子中,這個參數選取方法表現並不佳。我們提出另一個選取方法,在數值模擬上有較佳的表現。除了一般的均值估計,我們也成功地把此方法應用在群聚分析。 In 1995, Windham came out with an idea of weighted distribution in his thesis, Robustifying Model Fitting, and he used the idea to find a mean estimator when there are outliers in the original data. There is a tuning parameter in this estimator, and selecting the parameter will affect the mean estimate in the same data. In the same thesis, he also suggested a criterion of selecting the tuning parameter, but we found out that this criterion wasn’t doing well in some simulations. Considering the problem, we propose another criterion which can derive a better mean estimator. Besides, we can also apply this method to clustering problem. |
URI: | http://tdr.lib.ntu.edu.tw/handle/123456789/1203 |
DOI: | 10.6342/NTU201900783 |
Fulltext Rights: | 同意授權(全球公開) |
Appears in Collections: | 應用數學科學研究所 |
Files in This Item:
File | Size | Format | |
---|---|---|---|
ntu-107-1.pdf | 1.29 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.