Equity premium predictability using Bayesian regression mixtures

Xinxin Shang, Egon Kalotay

Research output: Chapter in Book/Report/Conference proceedingOther chapter contributionResearch

Abstract

This study investigates the predictability of equity returns in a mixture modelling framework that address several potential shortcomings of standard regression-based tools. We find that static single-equation predictive models are strongly rejected by the data in light of 2 and 3-component mixture regression specifications. We generalise standard regression mixtures to accommodate the possibility that predictive variables can be used to forecast time-variation in mixing weights. While within-sample variation in mixing weights appear forecastable, we find little benefit to forecasting such variation on an out-of-sample basis.
LanguageEnglish
Title of host publicationWorld finance conference venice, July 2 - 4, 2014 e-proceedings
PublisherWorld Finance Conference
Pages230-230
Number of pages1
ISBN (Print)9789899881617
Publication statusPublished - 2014
EventWorld finance conference - Venice
Duration: 2 Jul 20144 Jul 2014

Conference

ConferenceWorld finance conference
CityVenice
Period2/07/144/07/14

Fingerprint

Predictability
Equity premium
Modeling
Time variation
Equity returns

Cite this

Shang, X., & Kalotay, E. (2014). Equity premium predictability using Bayesian regression mixtures. In World finance conference venice, July 2 - 4, 2014 e-proceedings (pp. 230-230). World Finance Conference.
Shang, Xinxin ; Kalotay, Egon. / Equity premium predictability using Bayesian regression mixtures. World finance conference venice, July 2 - 4, 2014 e-proceedings. World Finance Conference, 2014. pp. 230-230
@inbook{a1033b98f0a245558edde5999acc04fd,
title = "Equity premium predictability using Bayesian regression mixtures",
abstract = "This study investigates the predictability of equity returns in a mixture modelling framework that address several potential shortcomings of standard regression-based tools. We find that static single-equation predictive models are strongly rejected by the data in light of 2 and 3-component mixture regression specifications. We generalise standard regression mixtures to accommodate the possibility that predictive variables can be used to forecast time-variation in mixing weights. While within-sample variation in mixing weights appear forecastable, we find little benefit to forecasting such variation on an out-of-sample basis.",
author = "Xinxin Shang and Egon Kalotay",
year = "2014",
language = "English",
isbn = "9789899881617",
pages = "230--230",
booktitle = "World finance conference venice, July 2 - 4, 2014 e-proceedings",
publisher = "World Finance Conference",

}

Shang, X & Kalotay, E 2014, Equity premium predictability using Bayesian regression mixtures. in World finance conference venice, July 2 - 4, 2014 e-proceedings. World Finance Conference, pp. 230-230, World finance conference, Venice, 2/07/14.

Equity premium predictability using Bayesian regression mixtures. / Shang, Xinxin; Kalotay, Egon.

World finance conference venice, July 2 - 4, 2014 e-proceedings. World Finance Conference, 2014. p. 230-230.

Research output: Chapter in Book/Report/Conference proceedingOther chapter contributionResearch

TY - CHAP

T1 - Equity premium predictability using Bayesian regression mixtures

AU - Shang, Xinxin

AU - Kalotay, Egon

PY - 2014

Y1 - 2014

N2 - This study investigates the predictability of equity returns in a mixture modelling framework that address several potential shortcomings of standard regression-based tools. We find that static single-equation predictive models are strongly rejected by the data in light of 2 and 3-component mixture regression specifications. We generalise standard regression mixtures to accommodate the possibility that predictive variables can be used to forecast time-variation in mixing weights. While within-sample variation in mixing weights appear forecastable, we find little benefit to forecasting such variation on an out-of-sample basis.

AB - This study investigates the predictability of equity returns in a mixture modelling framework that address several potential shortcomings of standard regression-based tools. We find that static single-equation predictive models are strongly rejected by the data in light of 2 and 3-component mixture regression specifications. We generalise standard regression mixtures to accommodate the possibility that predictive variables can be used to forecast time-variation in mixing weights. While within-sample variation in mixing weights appear forecastable, we find little benefit to forecasting such variation on an out-of-sample basis.

M3 - Other chapter contribution

SN - 9789899881617

SP - 230

EP - 230

BT - World finance conference venice, July 2 - 4, 2014 e-proceedings

PB - World Finance Conference

ER -

Shang X, Kalotay E. Equity premium predictability using Bayesian regression mixtures. In World finance conference venice, July 2 - 4, 2014 e-proceedings. World Finance Conference. 2014. p. 230-230