Testing for (in)finite moments

This paper proposes a test to verify whether the th moment of a random variable is finite. We use the fact that, under general assumptions, sample moments either converge to a finite number or diverge to infinity according as the corresponding population moment is finite or not. Building on this, we...

Full description

Bibliographic Details
Main Author: Trapani, Lorenzo
Format: Article
Published: Elsevier 2016
Subjects:
Online Access:https://eprints.nottingham.ac.uk/46950/
_version_ 1848797435184283648
author Trapani, Lorenzo
author_facet Trapani, Lorenzo
author_sort Trapani, Lorenzo
building Nottingham Research Data Repository
collection Online Access
description This paper proposes a test to verify whether the th moment of a random variable is finite. We use the fact that, under general assumptions, sample moments either converge to a finite number or diverge to infinity according as the corresponding population moment is finite or not. Building on this, we propose a test for the null that the th moment does not exist. Since, by construction, our test statistic diverges under the null and converges under the alternative, we propose a randomised testing procedure to discern between the two cases. We study the application of the test to raw data, and to regression residuals. Monte Carlo evidence shows that the test has the correct size and good power; the results are further illustrated through an application to financial data.
first_indexed 2025-11-14T20:03:50Z
format Article
id nottingham-46950
institution University of Nottingham Malaysia Campus
institution_category Local University
last_indexed 2025-11-14T20:03:50Z
publishDate 2016
publisher Elsevier
recordtype eprints
repository_type Digital Repository
spelling nottingham-469502020-05-04T20:03:54Z https://eprints.nottingham.ac.uk/46950/ Testing for (in)finite moments Trapani, Lorenzo This paper proposes a test to verify whether the th moment of a random variable is finite. We use the fact that, under general assumptions, sample moments either converge to a finite number or diverge to infinity according as the corresponding population moment is finite or not. Building on this, we propose a test for the null that the th moment does not exist. Since, by construction, our test statistic diverges under the null and converges under the alternative, we propose a randomised testing procedure to discern between the two cases. We study the application of the test to raw data, and to regression residuals. Monte Carlo evidence shows that the test has the correct size and good power; the results are further illustrated through an application to financial data. Elsevier 2016-03 Article PeerReviewed Trapani, Lorenzo (2016) Testing for (in)finite moments. Journal of Econometrics, 191 (1). pp. 57-68. ISSN 0304-4076 Finite moments; Randomised tests; Chover-type Law of the Iterated Logarithm; Strong Law of Large Numbers http://www.sciencedirect.com/science/article/pii/S0304407615002596 doi:10.1016/j.jeconom.2015.08.006 doi:10.1016/j.jeconom.2015.08.006
spellingShingle Finite moments; Randomised tests; Chover-type Law of the Iterated Logarithm; Strong Law of Large Numbers
Trapani, Lorenzo
Testing for (in)finite moments
title Testing for (in)finite moments
title_full Testing for (in)finite moments
title_fullStr Testing for (in)finite moments
title_full_unstemmed Testing for (in)finite moments
title_short Testing for (in)finite moments
title_sort testing for (in)finite moments
topic Finite moments; Randomised tests; Chover-type Law of the Iterated Logarithm; Strong Law of Large Numbers
url https://eprints.nottingham.ac.uk/46950/
https://eprints.nottingham.ac.uk/46950/
https://eprints.nottingham.ac.uk/46950/