Department of Economics and Business Economics

Complete subset regressions

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaperJournal articleResearchpeer-review

Standard

Complete subset regressions. / Elliott, G.; Gargano, A.; Timmermann, A.

In: Journal of Econometrics, Vol. 177, No. 2, 01.12.2013, p. 357-373.

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaperJournal articleResearchpeer-review

Harvard

Elliott, G, Gargano, A & Timmermann, A 2013, 'Complete subset regressions', Journal of Econometrics, vol. 177, no. 2, pp. 357-373. https://doi.org/10.1016/j.jeconom.2013.04.017

APA

Elliott, G., Gargano, A., & Timmermann, A. (2013). Complete subset regressions. Journal of Econometrics, 177(2), 357-373. https://doi.org/10.1016/j.jeconom.2013.04.017

CBE

Elliott G, Gargano A, Timmermann A. 2013. Complete subset regressions. Journal of Econometrics. 177(2):357-373. https://doi.org/10.1016/j.jeconom.2013.04.017

MLA

Elliott, G., A. Gargano, and A. Timmermann. "Complete subset regressions". Journal of Econometrics. 2013, 177(2). 357-373. https://doi.org/10.1016/j.jeconom.2013.04.017

Vancouver

Elliott G, Gargano A, Timmermann A. Complete subset regressions. Journal of Econometrics. 2013 Dec 1;177(2):357-373. https://doi.org/10.1016/j.jeconom.2013.04.017

Author

Elliott, G. ; Gargano, A. ; Timmermann, A. / Complete subset regressions. In: Journal of Econometrics. 2013 ; Vol. 177, No. 2. pp. 357-373.

Bibtex

@article{def21f421e474029a70e33f4bf4401e0,
title = "Complete subset regressions",
abstract = "This paper proposes a new method for combining forecasts based on complete subset regressions. For a given set of potential predictor variables we combine forecasts from all possible linear regression models that keep the number of predictors fixed. We explore how the choice of model complexity, as measured by the number of included predictor variables, can be used to trade off the bias and variance of the forecast errors, generating a setup akin to the efficient frontier known from modern portfolio theory. In an application to predictability of stock returns, we find that combinations of subset regressions can produce more accurate forecasts than conventional approaches based on equal-weighted forecasts (which fail to account for the dimensionality of the underlying models), combinations of univariate forecasts, or forecasts generated by methods such as bagging, ridge regression or Bayesian Model Averaging.",
author = "G. Elliott and A. Gargano and A. Timmermann",
year = "2013",
month = dec,
day = "1",
doi = "10.1016/j.jeconom.2013.04.017",
language = "English",
volume = "177",
pages = "357--373",
journal = "Journal of Econometrics",
issn = "0304-4076",
publisher = "Elsevier BV",
number = "2",

}

RIS

TY - JOUR

T1 - Complete subset regressions

AU - Elliott, G.

AU - Gargano, A.

AU - Timmermann, A.

PY - 2013/12/1

Y1 - 2013/12/1

N2 - This paper proposes a new method for combining forecasts based on complete subset regressions. For a given set of potential predictor variables we combine forecasts from all possible linear regression models that keep the number of predictors fixed. We explore how the choice of model complexity, as measured by the number of included predictor variables, can be used to trade off the bias and variance of the forecast errors, generating a setup akin to the efficient frontier known from modern portfolio theory. In an application to predictability of stock returns, we find that combinations of subset regressions can produce more accurate forecasts than conventional approaches based on equal-weighted forecasts (which fail to account for the dimensionality of the underlying models), combinations of univariate forecasts, or forecasts generated by methods such as bagging, ridge regression or Bayesian Model Averaging.

AB - This paper proposes a new method for combining forecasts based on complete subset regressions. For a given set of potential predictor variables we combine forecasts from all possible linear regression models that keep the number of predictors fixed. We explore how the choice of model complexity, as measured by the number of included predictor variables, can be used to trade off the bias and variance of the forecast errors, generating a setup akin to the efficient frontier known from modern portfolio theory. In an application to predictability of stock returns, we find that combinations of subset regressions can produce more accurate forecasts than conventional approaches based on equal-weighted forecasts (which fail to account for the dimensionality of the underlying models), combinations of univariate forecasts, or forecasts generated by methods such as bagging, ridge regression or Bayesian Model Averaging.

UR - http://www.scopus.com/inward/record.url?scp=84886722942&partnerID=8YFLogxK

U2 - 10.1016/j.jeconom.2013.04.017

DO - 10.1016/j.jeconom.2013.04.017

M3 - Journal article

AN - SCOPUS:84886722942

VL - 177

SP - 357

EP - 373

JO - Journal of Econometrics

JF - Journal of Econometrics

SN - 0304-4076

IS - 2

ER -