Asymptotics of forced-in variables from screening processes

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Freedman, Navidi and Peters (1988) had shown that when a two-stage variable selection procedure is used to screen linear regression equations, the type I error rate of the t-test for a forced-in-first estimator is inflated if the ratio of the number of parameters to number of data points is large. In this paper, we investigate the asymptotic distribution of the “force-in-last” estimator, and show that this estimator is bias toward zero asymptotically and its asymptotic standard error is close to the nominal standard error under certain conditions. To remedy the bias and reduce the inflation of the type I error rate, two estimators involved with propensity score (Rubin and Rosenbaum, 1983) were introduced.

Original languageEnglish
Pages (from-to)131-151
Number of pages21
JournalCommunications in Statistics - Theory and Methods
Volume24
Issue number1
DOIs
Publication statusPublished - 1995 Jan 1

Fingerprint

Screening
Estimator
Type I Error Rate
Standard error
Propensity Score
t-test
Selection Procedures
Variable Selection
Linear regression
Inflation
Asymptotic distribution
Categorical or nominal
Zero

All Science Journal Classification (ASJC) codes

  • Statistics and Probability

Cite this

@article{625a2c50384246f080a72e28c7234759,
title = "Asymptotics of forced-in variables from screening processes",
abstract = "Freedman, Navidi and Peters (1988) had shown that when a two-stage variable selection procedure is used to screen linear regression equations, the type I error rate of the t-test for a forced-in-first estimator is inflated if the ratio of the number of parameters to number of data points is large. In this paper, we investigate the asymptotic distribution of the “force-in-last” estimator, and show that this estimator is bias toward zero asymptotically and its asymptotic standard error is close to the nominal standard error under certain conditions. To remedy the bias and reduce the inflation of the type I error rate, two estimators involved with propensity score (Rubin and Rosenbaum, 1983) were introduced.",
author = "Iebin Lian",
year = "1995",
month = "1",
day = "1",
doi = "10.1080/03610929508831479",
language = "English",
volume = "24",
pages = "131--151",
journal = "Communications in Statistics - Theory and Methods",
issn = "0361-0926",
publisher = "Taylor and Francis Ltd.",
number = "1",

}

Asymptotics of forced-in variables from screening processes. / Lian, Iebin.

In: Communications in Statistics - Theory and Methods, Vol. 24, No. 1, 01.01.1995, p. 131-151.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Asymptotics of forced-in variables from screening processes

AU - Lian, Iebin

PY - 1995/1/1

Y1 - 1995/1/1

N2 - Freedman, Navidi and Peters (1988) had shown that when a two-stage variable selection procedure is used to screen linear regression equations, the type I error rate of the t-test for a forced-in-first estimator is inflated if the ratio of the number of parameters to number of data points is large. In this paper, we investigate the asymptotic distribution of the “force-in-last” estimator, and show that this estimator is bias toward zero asymptotically and its asymptotic standard error is close to the nominal standard error under certain conditions. To remedy the bias and reduce the inflation of the type I error rate, two estimators involved with propensity score (Rubin and Rosenbaum, 1983) were introduced.

AB - Freedman, Navidi and Peters (1988) had shown that when a two-stage variable selection procedure is used to screen linear regression equations, the type I error rate of the t-test for a forced-in-first estimator is inflated if the ratio of the number of parameters to number of data points is large. In this paper, we investigate the asymptotic distribution of the “force-in-last” estimator, and show that this estimator is bias toward zero asymptotically and its asymptotic standard error is close to the nominal standard error under certain conditions. To remedy the bias and reduce the inflation of the type I error rate, two estimators involved with propensity score (Rubin and Rosenbaum, 1983) were introduced.

UR - http://www.scopus.com/inward/record.url?scp=0038230302&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0038230302&partnerID=8YFLogxK

U2 - 10.1080/03610929508831479

DO - 10.1080/03610929508831479

M3 - Article

AN - SCOPUS:0038230302

VL - 24

SP - 131

EP - 151

JO - Communications in Statistics - Theory and Methods

JF - Communications in Statistics - Theory and Methods

SN - 0361-0926

IS - 1

ER -