1. NCCU Academic Hub
  2. 學術產出
  3. College of Commerce
  4. Periodical Articles
  5. Item

學術產出-Periodical Articles

Article View/Open

Publication Export

Google ScholarTM

政大圖書館

Citation Infomation

題名 Robust diagnostics for the heteroscedastic regression model
作者 Cheng, Tsung-Chi
鄭宗記
貢獻者 統計系
關鍵詞 Heteroscedasticity; Outlier; Residual maximum likelihood; Robust diagnostics; Search Algorithms; Trimmed likelihood; Estimation; Learning algorithms; Regression analysis; Maximum likelihood estimation
日期 2011-04
上傳時間 22-Jun-2015 16:04:30 (UTC+8)
摘要 The assumption of equal variance in the normal regression model is not always appropriate. Cook and Weisberg (1983) provide a score test to detect heteroscedasticity, while Patterson and Thompson (1971) propose the residual maximum likelihood (REML) estimation to estimate variance components in the context of an unbalanced incomplete-block design. REML is often preferred to the maximum likelihood estimation as a method of estimating covariance parameters in a linear model. However, outliers may have some effect on the estimate of the variance function. This paper incorporates the maximum trimming likelihood estimation (Hadi and Luceo, 1997; Vandev and Neykov, 1998) in REML to obtain a robust estimation of modelling variance heterogeneity. Both the forward search algorithm of Atkinson (1994) and the fast algorithm of Neykov et al. (2007) are employed to find the resulting estimator. Simulation and real data examples are used to illustrate the performance of the proposed approach. © 2010 Published by Elsevier B.V.
關聯 Computational Statistics and Data Analysis, 55(4), 1845-1866
資料類型 article
DOI http://dx.doi.org/10.1016/j.csda.2010.11.024
dc.contributor 統計系
dc.creator (作者) Cheng, Tsung-Chi
dc.creator (作者) 鄭宗記zh_TW
dc.date (日期) 2011-04
dc.date.accessioned 22-Jun-2015 16:04:30 (UTC+8)-
dc.date.available 22-Jun-2015 16:04:30 (UTC+8)-
dc.date.issued (上傳時間) 22-Jun-2015 16:04:30 (UTC+8)-
dc.identifier.uri (URI) http://nccur.lib.nccu.edu.tw/handle/140.119/76045-
dc.description.abstract (摘要) The assumption of equal variance in the normal regression model is not always appropriate. Cook and Weisberg (1983) provide a score test to detect heteroscedasticity, while Patterson and Thompson (1971) propose the residual maximum likelihood (REML) estimation to estimate variance components in the context of an unbalanced incomplete-block design. REML is often preferred to the maximum likelihood estimation as a method of estimating covariance parameters in a linear model. However, outliers may have some effect on the estimate of the variance function. This paper incorporates the maximum trimming likelihood estimation (Hadi and Luceo, 1997; Vandev and Neykov, 1998) in REML to obtain a robust estimation of modelling variance heterogeneity. Both the forward search algorithm of Atkinson (1994) and the fast algorithm of Neykov et al. (2007) are employed to find the resulting estimator. Simulation and real data examples are used to illustrate the performance of the proposed approach. © 2010 Published by Elsevier B.V.
dc.format.extent 634302 bytes-
dc.format.mimetype application/pdf-
dc.relation (關聯) Computational Statistics and Data Analysis, 55(4), 1845-1866
dc.subject (關鍵詞) Heteroscedasticity; Outlier; Residual maximum likelihood; Robust diagnostics; Search Algorithms; Trimmed likelihood; Estimation; Learning algorithms; Regression analysis; Maximum likelihood estimation
dc.title (題名) Robust diagnostics for the heteroscedastic regression model
dc.type (資料類型) articleen
dc.identifier.doi (DOI) 10.1016/j.csda.2010.11.024
dc.doi.uri (DOI) http://dx.doi.org/10.1016/j.csda.2010.11.024