Finite sample properties of tests based on prewhitened nonparametric covariance estimators

DOI10.1214/17-EJS1281zbMATH Open1365.62075arXiv1409.1419MaRDI QIDQ2396342FDOQ2396342

Authors: David Preinerstorfer 

Publication date: 8 June 2017

Published in: Electronic Journal of Statistics (Search for Journal in Brave)

Abstract: We analytically investigate size and power properties of a popular family of procedures for testing linear restrictions on the coefficient vector in a linear regression model with temporally dependent errors. The tests considered are autocorrelation-corrected F-type tests based on prewhitened nonparametric covariance estimators that possibly incorporate a data-dependent bandwidth parameter, e.g., estimators as considered in Andrews and Monahan (1992), Newey and West (1994), or Rho and Shao (2013). For design matrices that are generic in a measure theoretic sense we prove that these tests either suffer from extreme size distortions or from strong power deficiencies. Despite this negative result we demonstrate that a simple adjustment procedure based on artificial regressors can often resolve this problem.

Full work available at URL: https://arxiv.org/abs/1409.1419

Recommendations

• On size and power of heteroskedasticity and autocorrelation robust tests
• Semi-strong linearity testing in linear models with dependent but uncorrelated errors
• Finite sample power of linear regression autocorrelation tests
• MORE EFFICIENT ESTIMATION IN NONPARAMETRIC REGRESSION WITH NONPARAMETRIC AUTOCORRELATED ERRORS
• Simple Robust Testing of Hypotheses in Nonlinear Models

zbMATH Keywords

size distortionprewhiteningpower deficiencyartificial regressorsautocorrelation robustness

Mathematics Subject Classification ID

Parametric hypothesis testing (62F03) Linear regression; mixed models (62J05) Time series, auto-correlation, regression, etc. in statistics (GARCH) (62M10) Robustness and adaptive procedures (parametric inference) (62F35) Inference from stochastic processes and spectral analysis (62M15)

Cited In (10)

• Improving the bandwidth-free inference methods by prewhitening
• Should we go one step further? An accurate comparison of one-step and two-step procedures in a generalized method of moments framework
• HOW TO AVOID THE ZERO-POWER TRAP IN TESTING FOR CORRELATION
• Prewhitened long-run variance estimation robust to nonstationarity
• Controlling the size of autocorrelation robust tests
• Further results on size and power of heteroskedasticity and autocorrelation robust tests, with an application to trend testing
• On size and power of heteroskedasticity and autocorrelation robust tests
• Semi-strong linearity testing in linear models with dependent but uncorrelated errors
• A Modified Nonparametric Prewhitened Covariance Estimator
• Asymptotic \(F\) and \(t\) tests in an efficient GMM setting

This page was built for publication: Finite sample properties of tests based on prewhitened nonparametric covariance estimators

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2396342)