Forecasting stock returns with large dimensional factor models

Alessandro Giovannelli; Daniele Massacci; Stefano Soccorsi

doi:10.1016/j.jempfin.2021.07.009

Forecasting stock returns with large dimensional factor models

Alessandro Giovannelli, Daniele Massacci^*, Stefano Soccorsi

^*Corresponding author for this work

Banking & Finance

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

11 Citations (Scopus)

Abstract

We study equity premium out-of-sample predictability by extracting the information contained in a high number of macroeconomic predictors via large dimensional factor models. We compare the well-known factor model with a static representation of the common components with the Generalized Dynamic Factor Model, which accounts for time series dependence in the common components. Using statistical and economic evaluation criteria, we empirically show that the Generalized Dynamic Factor Model helps predicting the equity premium. Exploiting the link between business cycle and return predictability, we find accurate predictions also by combining rolling and recursive forecasts in real-time.

Original language	English
Pages (from-to)	252-269
Number of pages	18
Journal	Journal of Empirical Finance
Volume	63
DOIs	https://doi.org/10.1016/j.jempfin.2021.07.009
Publication status	Published - Sept 2021

Keywords

Factor model
Forecast evaluation
Large data sets
Stock returns forecasting

Access to Document

10.1016/j.jempfin.2021.07.009

Cite this

@article{634ca4c613904b098c6770d85759b7ba,

title = "Forecasting stock returns with large dimensional factor models",

abstract = "We study equity premium out-of-sample predictability by extracting the information contained in a high number of macroeconomic predictors via large dimensional factor models. We compare the well-known factor model with a static representation of the common components with the Generalized Dynamic Factor Model, which accounts for time series dependence in the common components. Using statistical and economic evaluation criteria, we empirically show that the Generalized Dynamic Factor Model helps predicting the equity premium. Exploiting the link between business cycle and return predictability, we find accurate predictions also by combining rolling and recursive forecasts in real-time.",

keywords = "Factor model, Forecast evaluation, Large data sets, Stock returns forecasting",

author = "Alessandro Giovannelli and Daniele Massacci and Stefano Soccorsi",

note = "Funding Information: ? This paper benefits from comments from participants at the CORE@50 Conference (Louvain-la-Neuve, Belgium), the 36th International Symposium on Forecasting (Santander, Spain) and the Macroeconomics and Financial Time Series Analysis workshop (Lancaster, UK) and from conversations with Marc Hallin and Marco Lippi. Errors and omissions are the authors? responsibility. Publisher Copyright: {\textcopyright} 2021 Elsevier B.V. Copyright: Copyright 2021 Elsevier B.V., All rights reserved.",

year = "2021",

month = sep,

doi = "10.1016/j.jempfin.2021.07.009",

language = "English",

volume = "63",

pages = "252--269",

journal = "Journal of Empirical Finance",

issn = "0927-5398",

publisher = "Elsevier",

}

TY - JOUR

T1 - Forecasting stock returns with large dimensional factor models

AU - Giovannelli, Alessandro

AU - Massacci, Daniele

AU - Soccorsi, Stefano

N1 - Funding Information: ? This paper benefits from comments from participants at the CORE@50 Conference (Louvain-la-Neuve, Belgium), the 36th International Symposium on Forecasting (Santander, Spain) and the Macroeconomics and Financial Time Series Analysis workshop (Lancaster, UK) and from conversations with Marc Hallin and Marco Lippi. Errors and omissions are the authors? responsibility. Publisher Copyright: © 2021 Elsevier B.V. Copyright: Copyright 2021 Elsevier B.V., All rights reserved.

PY - 2021/9

Y1 - 2021/9

N2 - We study equity premium out-of-sample predictability by extracting the information contained in a high number of macroeconomic predictors via large dimensional factor models. We compare the well-known factor model with a static representation of the common components with the Generalized Dynamic Factor Model, which accounts for time series dependence in the common components. Using statistical and economic evaluation criteria, we empirically show that the Generalized Dynamic Factor Model helps predicting the equity premium. Exploiting the link between business cycle and return predictability, we find accurate predictions also by combining rolling and recursive forecasts in real-time.

AB - We study equity premium out-of-sample predictability by extracting the information contained in a high number of macroeconomic predictors via large dimensional factor models. We compare the well-known factor model with a static representation of the common components with the Generalized Dynamic Factor Model, which accounts for time series dependence in the common components. Using statistical and economic evaluation criteria, we empirically show that the Generalized Dynamic Factor Model helps predicting the equity premium. Exploiting the link between business cycle and return predictability, we find accurate predictions also by combining rolling and recursive forecasts in real-time.

KW - Factor model

KW - Forecast evaluation

KW - Large data sets

KW - Stock returns forecasting

UR - http://www.scopus.com/inward/record.url?scp=85111214881&partnerID=8YFLogxK

U2 - 10.1016/j.jempfin.2021.07.009

DO - 10.1016/j.jempfin.2021.07.009

M3 - Article

AN - SCOPUS:85111214881

SN - 0927-5398

VL - 63

SP - 252

EP - 269

JO - Journal of Empirical Finance

JF - Journal of Empirical Finance

ER -

Forecasting stock returns with large dimensional factor models

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this