A Paradoxical Result in Estimating Regression Coefficients

被引：10

作者：

Tarpey, Thaddeus ^{[1
]}

Ogden, R. Todd ^{[2
]}

Petkova, Eva ^{[3
,4
]}

Christensen, Ronald ^{[5
]}

机构：

[1] Wright State Univ, Dept Math & Stat, Dayton, OH 45435 USA

[2] Columbia Univ, Dept Biostat, Mailman Sch Publ Hlth, New York, NY 10032 USA

[3] NYU, Dept Child & Adolescent Psychiat, New York, NY 10016 USA

[4] Nathan S Kline Inst Psychiat Res, Orangeburg, NY 10962 USA

[5] Univ New Mexico, Dept Math & Stat, Albuquerque, NM USA

来源：

AMERICAN STATISTICIAN | 2014年 / 68卷 / 04期

关键词：

Jensen's inequality; Ratio estimator; Pooled variance estimator; Moderator analysis;

D O I：

10.1080/00031305.2014.940467

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：

This article presents a counterintuitive result regarding the estimation of a regression slope coefficient. Paradoxically, the precision of the slope estimator can deteriorate when additional information is used to estimate its value. In a randomized experiment, the distribution of baseline variables should be identical across treatments due to randomization. The motivation for this article came from noting that the precision of slope estimators deteriorated when pooling baseline predictors across treatment groups.

引用

页码：271 / 276

页数：6

共 50 条

[41] ALTERNATIVE EXPLANATION OF A PARADOXICAL SKIN GRAFTING RESULT
AMOS, DB
TRANSPLANTATION, 1971, 12 (05) : 411 - &
[42] PATH COEFFICIENTS VERSUS REGRESSION COEFFICIENTS
BLALOCK, HM
AMERICAN JOURNAL OF SOCIOLOGY, 1967, 72 (06) : 675 - 676
[43] Estimation of regression coefficients of interest when other regression coefficients are of no interest
Magnus, JR
Durbin, J
ECONOMETRICA, 1999, 67 (03) : 639 - 643
[44] COMBINING OF INDEPENDENT REGRESSION ESTIMATES IS NOT PARADOXICAL
MANTEL, N
BROWN, CC
AMERICAN STATISTICIAN, 1973, 27 (03): : 125 - 126
[45] ESTIMATING EQUATION COEFFICIENTS
MADDOX, RN
PETERSON, ER
CHEMICAL ENGINEERING, 1986, 93 (04) : 127 - 129
[46] Estimating competition coefficients from census data: A re-examination of the regression technique.
Fox, BJ
Luo, J
OIKOS, 1996, 77 (02) : 291 - 300
[47] Multiple imputation methods for estimating regression coefficients in the competing risks model with missing cause of failure
Lu, KF
Tsiatis, AA
BIOMETRICS, 2001, 57 (04) : 1191 - 1197
[48] Comparing Regression Coefficients Between Nested Linear Models for Clustered Data With Generalized Estimating Equations
Yan, Jun
Aseltine, Robert H., Jr.
Harel, Ofer
JOURNAL OF EDUCATIONAL AND BEHAVIORAL STATISTICS, 2013, 38 (02) : 172 - 189
[49] Estimating the standardized regression coefficients of design variables in daylighting and energy performance of buildings in the face of multicollinearity
Allam, Amr S.
Bassioni, Hesham A.
Kamel, Wael
Ayoub, Mohammed
SOLAR ENERGY, 2020, 211 : 1184 - 1193
[50] ESTIMATION OF REGRESSION COEFFICIENTS
ALAM, K
HAWKES, JS
SCANDINAVIAN JOURNAL OF STATISTICS, 1978, 5 (03) : 169 - 172

← 1 2 3 4 5 →