On false discoveries of standard t-tests in investment management applications

Financial managers routinely use the one-sample t-test to evaluate whether the mean returns of investment assets, strategies or funds are significantly different from zero. Simultaneously, however, they often ignore the fact that its application is not generally justified, in other words, that its u...

Full description

Saved in:
Bibliographic Details
Published inReview of managerial science Vol. 16; no. 3; pp. 751 - 768
Main Author Auer, Benjamin R.
Format Journal Article
LanguageEnglish
Published Berlin/Heidelberg Springer Berlin Heidelberg 01.04.2022
Springer Nature B.V
Subjects
Accounting/Auditing
Banking
Business and Management
Business Strategy/Leadership
Marketing
Monte Carlo simulation
Original Paper
Student's t-test
Outliers
Normal quantile
C12
C15
G11
One-sample t-test
Autocorrelation
Non-normality
Online AccessGet full text

Cover

More Information
Summary:Financial managers routinely use the one-sample t-test to evaluate whether the mean returns of investment assets, strategies or funds are significantly different from zero. Simultaneously, however, they often ignore the fact that its application is not generally justified, in other words, that its usefulness depends on the properties of the population. We show by Monte Carlo simulation that, especially in skewed and/or autocorrelated populations, test decisions based on the t-test can be severely biased. More specifically, for sample sizes typically used in investment performance evaluation, the probability of falsely diagnosing a significant mean return–the false discovery rate–is significantly higher than the nominal error probability set in testing. We additionally illustrate that the popular empirical practices of (i) replacing the t-quantile with the standard normal quantile and (ii) removing outliers before conducting the t-test have crucial elevating impact on the false discovery rate.
ISSN:1863-6683
1863-6691
DOI:10.1007/s11846-021-00453-0