Predicting corporate policies using downside risk: A machine learning approach

Doron Avramov; Minwen Li; Hao Wang

doi:10.1016/j.jempfin.2021.04.009

Predicting corporate policies using downside risk: A machine learning approach

Doron Avramov, Minwen Li, Hao Wang^*

^*Corresponding author for this work

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

2 Scopus citations

Abstract

This paper develops a text-based downside risk measure using corporate annual reports and assesses its ability to forecast future corporate policies. The forward-looking measure dynamically captures adverse firm conditions evolving from economic fundamentals. When the measure is below its sample average, leverage, investment, R&D, employment, and dividends consistently fall. When the measure rises, firms increase cash holdings. The proposed measure also delivers robust and persistent forecasts based on in-sample and out-of-sample LASSO regressions.

Original language	American English
Pages (from-to)	1-26
Number of pages	26
Journal	Journal of Empirical Finance
Volume	63
DOIs	https://doi.org/10.1016/j.jempfin.2021.04.009
State	Published - Sep 2021
Externally published	Yes

Bibliographical note

Funding Information:
We thank Jeawon Choi, Ran Duchin, Campbell Harvey, Gerald Hoberg, Si Li, Nagpurnanand Prabhala, Gordon Phillips, Yongxiang Wang, Xuan Tian, Jianfeng Yu, Lihong Zhang, Hao Zhou, and seminar participants at Tsinghua Finance Group Workshop 2013, PBC School of Finance at Tsinghua University, Xiamen University, Peking University, Summer Institute of Finance 2015 by CKGSB, Shanghai Advanced Institute of Finance (SAIF) and China International Conference in Finance (2016) for their comments. Hao Wang acknowledges financial support from Tsinghua University Research Grant (Grant No. 2019THZWLJ14). Minwen Li acknowledges financial support from the National Natural Science Foundation of China (Grant No. 71402078).

Funding Information:
We thank Jeawon Choi, Ran Duchin, Campbell Harvey, Gerald Hoberg, Si Li, Nagpurnanand Prabhala, Gordon Phillips, Yongxiang Wang, Xuan Tian, Jianfeng Yu, Lihong Zhang, Hao Zhou, and seminar participants at Tsinghua Finance Group Workshop 2013, PBC School of Finance at Tsinghua University, Xiamen University, Peking University, Summer Institute of Finance 2015 by CKGSB, Shanghai Advanced Institute of Finance (SAIF) and China International Conference in Finance (2016) for their comments. Hao Wang acknowledges financial support from Tsinghua University Research Grant (Grant No. 2019THZWLJ14 ). Minwen Li acknowledges financial support from the National Natural Science Foundation of China (Grant No. 71402078 ).

Publisher Copyright:
© 2021 Elsevier B.V.

Access to Document

10.1016/j.jempfin.2021.04.009

Cite this

@article{928ee4e248e046feb2f38d54e8a5115f,

title = "Predicting corporate policies using downside risk: A machine learning approach",

abstract = "This paper develops a text-based downside risk measure using corporate annual reports and assesses its ability to forecast future corporate policies. The forward-looking measure dynamically captures adverse firm conditions evolving from economic fundamentals. When the measure is below its sample average, leverage, investment, R&D, employment, and dividends consistently fall. When the measure rises, firms increase cash holdings. The proposed measure also delivers robust and persistent forecasts based on in-sample and out-of-sample LASSO regressions.",

author = "Doron Avramov and Minwen Li and Hao Wang",

note = "Funding Information: We thank Jeawon Choi, Ran Duchin, Campbell Harvey, Gerald Hoberg, Si Li, Nagpurnanand Prabhala, Gordon Phillips, Yongxiang Wang, Xuan Tian, Jianfeng Yu, Lihong Zhang, Hao Zhou, and seminar participants at Tsinghua Finance Group Workshop 2013, PBC School of Finance at Tsinghua University, Xiamen University, Peking University, Summer Institute of Finance 2015 by CKGSB, Shanghai Advanced Institute of Finance (SAIF) and China International Conference in Finance (2016) for their comments. Hao Wang acknowledges financial support from Tsinghua University Research Grant (Grant No. 2019THZWLJ14). Minwen Li acknowledges financial support from the National Natural Science Foundation of China (Grant No. 71402078). Funding Information: We thank Jeawon Choi, Ran Duchin, Campbell Harvey, Gerald Hoberg, Si Li, Nagpurnanand Prabhala, Gordon Phillips, Yongxiang Wang, Xuan Tian, Jianfeng Yu, Lihong Zhang, Hao Zhou, and seminar participants at Tsinghua Finance Group Workshop 2013, PBC School of Finance at Tsinghua University, Xiamen University, Peking University, Summer Institute of Finance 2015 by CKGSB, Shanghai Advanced Institute of Finance (SAIF) and China International Conference in Finance (2016) for their comments. Hao Wang acknowledges financial support from Tsinghua University Research Grant (Grant No. 2019THZWLJ14 ). Minwen Li acknowledges financial support from the National Natural Science Foundation of China (Grant No. 71402078 ). Publisher Copyright: {\textcopyright} 2021 Elsevier B.V.",

year = "2021",

month = sep,

doi = "10.1016/j.jempfin.2021.04.009",

language = "American English",

volume = "63",

pages = "1--26",

journal = "Journal of Empirical Finance",

issn = "0927-5398",

publisher = "Elsevier",

}

TY - JOUR

T1 - Predicting corporate policies using downside risk

T2 - A machine learning approach

AU - Avramov, Doron

AU - Li, Minwen

AU - Wang, Hao

N1 - Funding Information: We thank Jeawon Choi, Ran Duchin, Campbell Harvey, Gerald Hoberg, Si Li, Nagpurnanand Prabhala, Gordon Phillips, Yongxiang Wang, Xuan Tian, Jianfeng Yu, Lihong Zhang, Hao Zhou, and seminar participants at Tsinghua Finance Group Workshop 2013, PBC School of Finance at Tsinghua University, Xiamen University, Peking University, Summer Institute of Finance 2015 by CKGSB, Shanghai Advanced Institute of Finance (SAIF) and China International Conference in Finance (2016) for their comments. Hao Wang acknowledges financial support from Tsinghua University Research Grant (Grant No. 2019THZWLJ14). Minwen Li acknowledges financial support from the National Natural Science Foundation of China (Grant No. 71402078). Funding Information: We thank Jeawon Choi, Ran Duchin, Campbell Harvey, Gerald Hoberg, Si Li, Nagpurnanand Prabhala, Gordon Phillips, Yongxiang Wang, Xuan Tian, Jianfeng Yu, Lihong Zhang, Hao Zhou, and seminar participants at Tsinghua Finance Group Workshop 2013, PBC School of Finance at Tsinghua University, Xiamen University, Peking University, Summer Institute of Finance 2015 by CKGSB, Shanghai Advanced Institute of Finance (SAIF) and China International Conference in Finance (2016) for their comments. Hao Wang acknowledges financial support from Tsinghua University Research Grant (Grant No. 2019THZWLJ14 ). Minwen Li acknowledges financial support from the National Natural Science Foundation of China (Grant No. 71402078 ). Publisher Copyright: © 2021 Elsevier B.V.

PY - 2021/9

Y1 - 2021/9

N2 - This paper develops a text-based downside risk measure using corporate annual reports and assesses its ability to forecast future corporate policies. The forward-looking measure dynamically captures adverse firm conditions evolving from economic fundamentals. When the measure is below its sample average, leverage, investment, R&D, employment, and dividends consistently fall. When the measure rises, firms increase cash holdings. The proposed measure also delivers robust and persistent forecasts based on in-sample and out-of-sample LASSO regressions.

AB - This paper develops a text-based downside risk measure using corporate annual reports and assesses its ability to forecast future corporate policies. The forward-looking measure dynamically captures adverse firm conditions evolving from economic fundamentals. When the measure is below its sample average, leverage, investment, R&D, employment, and dividends consistently fall. When the measure rises, firms increase cash holdings. The proposed measure also delivers robust and persistent forecasts based on in-sample and out-of-sample LASSO regressions.

UR - http://www.scopus.com/inward/record.url?scp=85106963110&partnerID=8YFLogxK

U2 - 10.1016/j.jempfin.2021.04.009

DO - 10.1016/j.jempfin.2021.04.009

M3 - Article

AN - SCOPUS:85106963110

SN - 0927-5398

VL - 63

SP - 1

EP - 26

JO - Journal of Empirical Finance

JF - Journal of Empirical Finance

ER -

Predicting corporate policies using downside risk: A machine learning approach

Abstract

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this