Least product relative error estimation

JOURNAL OF MULTIVARIATE ANALYSIS

卷:144

Article

Chen, Kani¹  Lin, Yuanyuan²  Wang, Zhanfeng³  Ying, Zhiliang⁴ 

[1] Hong Kong Univ Sci & Technol, Dept Math, Kowloon, Hong Kong, Peoples R China

[2] Chinese Univ Hong Kong, Dept Stat, Shatin, Hong Kong, Peoples R China

[3] Univ Sci & Technol China, Dept Stat & Finance, Hefei 230026, Peoples R China

[4] Columbia Univ, Dept Stat, New York, NY 10027 USA

关键词: Multiplicative regression model;    Product form;    Relative error;    Scale invariance;    Variance estimation;   

DOI  :  10.1016/j.jmva.2015.10.017

来源: Elsevier

【 摘 要 】

A least product relative error criterion is proposed for multiplicative regression models. It is invariant under scale transformation of the outcome and covariates. In addition, the objective function is smooth and convex, resulting in a simple and uniquely defined estimator of the regression parameter. It is shown that the estimator is asymptotically normal and that the simple plug-in variance estimation is valid. Simulation results confirm that the proposed method performs well. An application to body fat calculation is presented to illustrate the new method. (C) 2015 Elsevier Inc. All rights reserved.

【 授权许可】

Free