Finite-sample exact tests for linear regressions with bounded dependent variables

Olivier Gossner; Karl H. Schlag

doi:10.1016/j.jeconom.2013.06.003

Finite-sample exact tests for linear regressions with bounded dependent variables

Olivier Gossner
, Karl H. Schlag

Research output: Contribution to journal › Article › peer-review

Abstract

We introduce tests for finite-sample linear regressions with heteroskedastic errors. The tests are exact, i.e., they have guaranteed type I error probabilities when bounds are known on the range of the dependent variable, without any assumptions about the noise structure. We provide upper bounds on probability of type II errors, and apply the tests to empirical data.

Original language	English
Pages (from-to)	75-84
Number of pages	10
Journal	Journal of Econometrics
Volume	177
Issue number	1
DOIs	https://doi.org/10.1016/j.jeconom.2013.06.003
Publication status	Published - 1 Jan 2013
Externally published	Yes

Keywords

Exact test
Heteroskedasticity
Nonparametric linear regression

Access to Document

10.1016/j.jeconom.2013.06.003

Cite this

@article{2e5b77a78f83479fa362ba8b0ce11930,

title = "Finite-sample exact tests for linear regressions with bounded dependent variables",

abstract = "We introduce tests for finite-sample linear regressions with heteroskedastic errors. The tests are exact, i.e., they have guaranteed type I error probabilities when bounds are known on the range of the dependent variable, without any assumptions about the noise structure. We provide upper bounds on probability of type II errors, and apply the tests to empirical data.",

keywords = "Exact test, Heteroskedasticity, Nonparametric linear regression",

author = "Olivier Gossner and Schlag, \{Karl H.\}",

year = "2013",

month = jan,

day = "1",

doi = "10.1016/j.jeconom.2013.06.003",

language = "English",

volume = "177",

pages = "75--84",

journal = "Journal of Econometrics",

issn = "0304-4076",

number = "1",

}

TY - JOUR

T1 - Finite-sample exact tests for linear regressions with bounded dependent variables

AU - Gossner, Olivier

AU - Schlag, Karl H.

PY - 2013/1/1

Y1 - 2013/1/1

N2 - We introduce tests for finite-sample linear regressions with heteroskedastic errors. The tests are exact, i.e., they have guaranteed type I error probabilities when bounds are known on the range of the dependent variable, without any assumptions about the noise structure. We provide upper bounds on probability of type II errors, and apply the tests to empirical data.

AB - We introduce tests for finite-sample linear regressions with heteroskedastic errors. The tests are exact, i.e., they have guaranteed type I error probabilities when bounds are known on the range of the dependent variable, without any assumptions about the noise structure. We provide upper bounds on probability of type II errors, and apply the tests to empirical data.

KW - Exact test

KW - Heteroskedasticity

KW - Nonparametric linear regression

U2 - 10.1016/j.jeconom.2013.06.003

DO - 10.1016/j.jeconom.2013.06.003

M3 - Article

AN - SCOPUS:84883448973

SN - 0304-4076

VL - 177

SP - 75

EP - 84

JO - Journal of Econometrics

JF - Journal of Econometrics

IS - 1

ER -

Finite-sample exact tests for linear regressions with bounded dependent variables

Abstract

Keywords

Access to Document

Fingerprint

Cite this