Home > Research > Publications & Outputs > Improved inference in regression with overlappi...
View graph of relations

Improved inference in regression with overlapping observations

Research output: Contribution to journalJournal article

Published

Standard

Improved inference in regression with overlapping observations. / Britten-Jones, Mark; Neuberger, Anthony; Nolte, Ingmar.

In: Journal of Business Finance and Accounting, Vol. 38, No. 5-6, 06.2011, p. 657-683.

Research output: Contribution to journalJournal article

Harvard

Britten-Jones, M, Neuberger, A & Nolte, I 2011, 'Improved inference in regression with overlapping observations', Journal of Business Finance and Accounting, vol. 38, no. 5-6, pp. 657-683. https://doi.org/10.1111/j.1468-5957.2011.02244.x

APA

Britten-Jones, M., Neuberger, A., & Nolte, I. (2011). Improved inference in regression with overlapping observations. Journal of Business Finance and Accounting, 38(5-6), 657-683. https://doi.org/10.1111/j.1468-5957.2011.02244.x

Vancouver

Britten-Jones M, Neuberger A, Nolte I. Improved inference in regression with overlapping observations. Journal of Business Finance and Accounting. 2011 Jun;38(5-6):657-683. https://doi.org/10.1111/j.1468-5957.2011.02244.x

Author

Britten-Jones, Mark ; Neuberger, Anthony ; Nolte, Ingmar. / Improved inference in regression with overlapping observations. In: Journal of Business Finance and Accounting. 2011 ; Vol. 38, No. 5-6. pp. 657-683.

Bibtex

@article{f8ab2317741648d09a6aeac8a5f5cdbf,
title = "Improved inference in regression with overlapping observations",
abstract = "We present an improved method for inference in linear regressions with overlapping observations. By aggregating the matrix of explanatory variables in a simple way, our method transforms the original regression into an equivalent representation in which the dependent variables are non-overlapping. This transformation removes that part of the autocorrelation in the error terms which is induced by the overlapping scheme. Our method can easily be applied within standard software packages since conventional inference procedures (OLS-, White-, Newey-West- standard errors) are asymptotically valid when applied to the transformed regression. Through Monte Carlo analysis we show that it performs better in finite samples than the methods applied to the original regression that are in common usage. We illustrate the significance of our method with three empirical applications.",
keywords = "long horizon, stock return predictability, induced autocorrelation",
author = "Mark Britten-Jones and Anthony Neuberger and Ingmar Nolte",
year = "2011",
month = jun
doi = "10.1111/j.1468-5957.2011.02244.x",
language = "English",
volume = "38",
pages = "657--683",
journal = "Journal of Business Finance and Accounting",
issn = "0306-686X",
publisher = "Wiley-Blackwell",
number = "5-6",

}

RIS

TY - JOUR

T1 - Improved inference in regression with overlapping observations

AU - Britten-Jones, Mark

AU - Neuberger, Anthony

AU - Nolte, Ingmar

PY - 2011/6

Y1 - 2011/6

N2 - We present an improved method for inference in linear regressions with overlapping observations. By aggregating the matrix of explanatory variables in a simple way, our method transforms the original regression into an equivalent representation in which the dependent variables are non-overlapping. This transformation removes that part of the autocorrelation in the error terms which is induced by the overlapping scheme. Our method can easily be applied within standard software packages since conventional inference procedures (OLS-, White-, Newey-West- standard errors) are asymptotically valid when applied to the transformed regression. Through Monte Carlo analysis we show that it performs better in finite samples than the methods applied to the original regression that are in common usage. We illustrate the significance of our method with three empirical applications.

AB - We present an improved method for inference in linear regressions with overlapping observations. By aggregating the matrix of explanatory variables in a simple way, our method transforms the original regression into an equivalent representation in which the dependent variables are non-overlapping. This transformation removes that part of the autocorrelation in the error terms which is induced by the overlapping scheme. Our method can easily be applied within standard software packages since conventional inference procedures (OLS-, White-, Newey-West- standard errors) are asymptotically valid when applied to the transformed regression. Through Monte Carlo analysis we show that it performs better in finite samples than the methods applied to the original regression that are in common usage. We illustrate the significance of our method with three empirical applications.

KW - long horizon

KW - stock return predictability

KW - induced autocorrelation

U2 - 10.1111/j.1468-5957.2011.02244.x

DO - 10.1111/j.1468-5957.2011.02244.x

M3 - Journal article

VL - 38

SP - 657

EP - 683

JO - Journal of Business Finance and Accounting

JF - Journal of Business Finance and Accounting

SN - 0306-686X

IS - 5-6

ER -