Empirical Decisions and Replicating Anomalies

52 Pages Posted: 13 Aug 2023

See all articles by Jiaqi Guo

Jiaqi Guo

Birmingham Business School

Peng Li

University of Bath, School of management

Date Written: August 11, 2023

Abstract

There is an ongoing debate about the reproducibility of anomalies and p-hacking (data mining) of anomaly discoveries. This paper simulates and evaluates the impact of empirical decisions on anomaly replication and p-hacking. To better capture the true anomaly effect, we aggregate its return across 96 sets of portfolio construction designs, avoiding dependence on any particular design. We develop a two-stage bootstrap approach to account for both sampling and empirical design variations and show that 70% of the published 173 anomalies can be replicated. Furthermore, we simulate anomaly discoveries through p-hacking activities and publication bias behavior. The findings indicate the existence of p-hacking attempts especially when the t-value threshold is 2 but the extent of p-hacking is not severe in anomaly studies.

Keywords: Empirical decisions, Anomaly replication, Bootstrap simulation, p-hacking, Data mining

JEL Classification: C58, G10, G11, G12

Suggested Citation

Guo, Jiaqi and Li, Peng, Empirical Decisions and Replicating Anomalies (August 11, 2023). Available at SSRN: https://ssrn.com/abstract=4538764

Jiaqi Guo (Contact Author)

Birmingham Business School

Peng Li

University of Bath, School of management ( email )

Claverton Down
Bath, BA2 7AY

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
31
Abstract Views
198
PlumX Metrics