An analytical shrinkage estimator for linear regression

被引：1

作者：

Lassance, Nathan ^{[1
]}

机构：

[1] UCLouvain, LFIN, LIDAM, Chaussee Binche 151, B-5000 Mons, Belgium

来源：

STATISTICS & PROBABILITY LETTERS | 2023年 / 194卷

关键词：

Linear regression; Prediction error; Shrinkage; Out-of-sample; SELECTION;

D O I：

10.1016/j.spl.2022.109760

中图分类号：

O21 [概率论与数理统计]; C8 [统计学];

学科分类号：

020208 ; 070103 ; 0714 ;

摘要：

We derive an analytical solution to the optimal shrinkage of OLS regression coefficients toward a constant target, under any first two moments of predictors. The estimator closely mimics the prediction performance of ridge penalty, which admits no general analytical solution.(c) 2022 Elsevier B.V. All rights reserved.

引用

页数：9

共 14 条

[1]

Ahmed S.E., 2014, Penalty, Shrinkage and Pretest Strategies: Variable Selection and Estimation, V1st ed., DOI 10.1007/978-3-319-03149-1

[2] PROPERTIES OF THE INVERSE OF A NONCENTRAL WISHART MATRIX [J].

Hillier, Grant ;

Kan, Raymond .

ECONOMETRIC THEORY, 2022, 38 (06) :1092-1116

[3] RIDGE REGRESSION - BIASED ESTIMATION FOR NONORTHOGONAL PROBLEMS [J].

HOERL, AE ;

KENNARD, RW .

TECHNOMETRICS, 1970, 12 (01) :55-&

[4] Optimal portfolio choice with parameter uncertainty [J].

Kan, Raymond ;

Zhou, Guofu .

JOURNAL OF FINANCIAL AND QUANTITATIVE ANALYSIS, 2007, 42 (03) :621-656

[5]

Lassance N., 2022, 3855546 SSRN

[6]

Lederer J., 2021, Fundamentals of High-Dimensional Statistics: With Exercises and R Labs. Springer Texts in Statistics

[7]

Lederer J, 2015, AAAI CONF ARTIF INTE, P2729

[8]

Nakkiran P., 2021, INT C LEARNING REPRE

[9] RANDOM CALIBRATION WITH MANY MEASUREMENTS - AN APPLICATION OF STEIN ESTIMATION [J].

OMAN, SD .

TECHNOMETRICS, 1991, 33 (02) :187-195

[10] Penalty, post pretest and shrinkage strategies in a partially linear model [J].

Phukongtong, Siwaporn ;

Lisawadi, Supranee ;

Ahmed, S. Ejaz .

COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION, 2022, 51 (10) :6004-6025

← 1 2 →