Causal factor investing : can factor investing become scientific? / Marcos M. López de Prado, ADIA Lab.

Format:

Book

Author/Creator:

López de Prado, Marcos Mailoc, author.

Series:

Cambridge elements. Elements in quantitative finance, 2631-8571.

Cambridge elements. Elements in quantitative finance, 2631-8571

Language:

English

Subjects (All):

Investments.

Asset allocation.

Portfolio management.

Physical Description:

1 online resource (81 pages) : digital, PDF file(s).

Edition:

First edition.

Place of Publication:

Cambridge, United Kingdom ; New York, NY : Cambridge University Press, 2023.

Summary:

Virtually all journal articles in the factor investing literature make associational claims, instead of causal claims. Authors do not identify the causal graph consistent with the observed phenomenon, they justify their chosen model specification in terms of correlations, and they do not propose experiments for falsifying causal mechanisms. Absent a causal theory, their findings are likely false, due to rampant backtest overfitting and incorrect specification choices. This Element differentiates between type-A and type-B spurious claims, and explains how both types prevent factor investing from advancing beyond its current phenomenological stage. It analyzes the current state of causal confusion in the factor investing literature, and proposes solutions with the potential to transform factor investing into a truly scientific discipline. This title is also available as Open Access on Cambridge Core.

Contents:

Cover

Title page

Copyright page

Causal Factor Investing: Can Factor Investing Become Scientific?

Contents

1 Introduction

2 Association vs Causation

3 The Three Steps of Scientific Discovery

3.1 The Phenomenological Step

3.2 The Theoretical Step

3.3 The Falsification Step

3.4 Demarcation and Falsificationism in Statistics

4 Causal Inference

4.1 Interventional Studies

4.2 Natural Experiments

4.3 Simulated Interventions

4.3.1 Causal Discovery

4.3.2 Do-Calculus

4.3.2.1 Blocked Paths

4.3.2.2 Backdoor Adjustment

4.3.2.3 Front-Door Adjustment

4.3.2.4 Instrumental Variables

5 Causality in Econometrics

5.1 Authors often Mistake Causality for Association

5.2 Authors often Misunderstand the Meaning of β

5.3 Authors Often Mistake Association for Causality

6 Causality in Factor Investing

6.1 Causal Content

6.2 Omitted Mediation Analysis

6.2.1 Example of Factor Causal Mechanism

6.3 Causal Denial

6.4 Spurious Investment Factors

6.4.1 Type-A Spuriosity

6.4.1.1 P-Hacking

6.4.1.2 Backtest Overfitting

6.4.2 Type-B Spuriosity

6.4.2.1 Under-Controlling

6.4.2.2 Over-Controlling

6.4.2.3 Specification-Searching

6.4.2.4 Failure to Account for Temporal Properties

6.5 Hierarchy of Evidence

7 Monte Carlo Experiments

7.1 Fork

7.2 Immorality

7.3 Chain

7.4 An Alternative Explanation for Factors

8 Conclusions

8.1 Factor Investing Is a Black-Box

8.2 The Economic Incentives for Associational Studies

8.3 The Dawn of Causal Factor Investing

Appendix

A.1 Proof of Proposition in Section 6.4.2.1

A.2 Proof of Proposition in Section 7.2

B.1 Code for Experiment in Section 7.1

B.2 Code for Experiment in Section 7.2

B.3 Code for Experiment in Section 7.3

References

Acknowledgments

About the Author

About ADIA Lab.

Notes:

Open Access

Title from publisher's bibliographic system (viewed on 12 Oct 2023).

ISBN:

1-009-39730-3

1-009-39732-X

1-009-39731-1