Influence Analysis of Nongaussianity by Applying Projection Pursuit

碩士 === 國立中正大學 === 統計科學所 === 94 === Gaussian distribution is the least structured from the information-theoretic point of view. In this thesis, the projection pursuit is performed by finding the most nongaussian projection to explore the clustering structure of the data. We use kurtosis as a measure...

Full description

Bibliographic Details
Main Authors: Chin-Zen Cheng, 鄭清仁
Other Authors: none
Format: Others
Language:en_US
Published: 2006
Online Access:http://ndltd.ncl.edu.tw/handle/53766687957568781608
id ndltd-TW-094CCU05337013
record_format oai_dc
spelling ndltd-TW-094CCU053370132015-10-13T10:45:17Z http://ndltd.ncl.edu.tw/handle/53766687957568781608 Influence Analysis of Nongaussianity by Applying Projection Pursuit Chin-Zen Cheng 鄭清仁 碩士 國立中正大學 統計科學所 94 Gaussian distribution is the least structured from the information-theoretic point of view. In this thesis, the projection pursuit is performed by finding the most nongaussian projection to explore the clustering structure of the data. We use kurtosis as a measure of nongaussianity to find the projection direction. Kurtosis is well known to be sensitive to abnormal observations, henceforth the projection direction will be essentially affected by unusual points. The perturbation theory provides a useful tool in sensitivity analysis. In this thesis, we develop influence functions for the projection direction to investigate the influence of unusual observations. It is well-known that single-perturbation diagnostics can suffer from the masking effect. Hence we also develop the pair-perturbation influence functions to detect the masked influential points and outliers. A simulated data and a specific data example are provided to illustrate the applications of these approaches. none 黃郁芬 2006 學位論文 ; thesis 41 en_US
collection NDLTD
language en_US
format Others
sources NDLTD
description 碩士 === 國立中正大學 === 統計科學所 === 94 === Gaussian distribution is the least structured from the information-theoretic point of view. In this thesis, the projection pursuit is performed by finding the most nongaussian projection to explore the clustering structure of the data. We use kurtosis as a measure of nongaussianity to find the projection direction. Kurtosis is well known to be sensitive to abnormal observations, henceforth the projection direction will be essentially affected by unusual points. The perturbation theory provides a useful tool in sensitivity analysis. In this thesis, we develop influence functions for the projection direction to investigate the influence of unusual observations. It is well-known that single-perturbation diagnostics can suffer from the masking effect. Hence we also develop the pair-perturbation influence functions to detect the masked influential points and outliers. A simulated data and a specific data example are provided to illustrate the applications of these approaches.
author2 none
author_facet none
Chin-Zen Cheng
鄭清仁
author Chin-Zen Cheng
鄭清仁
spellingShingle Chin-Zen Cheng
鄭清仁
Influence Analysis of Nongaussianity by Applying Projection Pursuit
author_sort Chin-Zen Cheng
title Influence Analysis of Nongaussianity by Applying Projection Pursuit
title_short Influence Analysis of Nongaussianity by Applying Projection Pursuit
title_full Influence Analysis of Nongaussianity by Applying Projection Pursuit
title_fullStr Influence Analysis of Nongaussianity by Applying Projection Pursuit
title_full_unstemmed Influence Analysis of Nongaussianity by Applying Projection Pursuit
title_sort influence analysis of nongaussianity by applying projection pursuit
publishDate 2006
url http://ndltd.ncl.edu.tw/handle/53766687957568781608
work_keys_str_mv AT chinzencheng influenceanalysisofnongaussianitybyapplyingprojectionpursuit
AT zhèngqīngrén influenceanalysisofnongaussianitybyapplyingprojectionpursuit
_version_ 1716833019575140352