A randomised sequential procedure to determine the number of factors

This paper proposes a procedure to estimate the number of common factors k in a static approximate factor model. The building block of the analysis is the fact that the first k eigenvalues of the covariance matrix of the data diverge, whilst the others stay bounded. On the grounds of this, we propos...

Full description

Bibliographic Details
Main Author: Trapani, Lorenzo
Format: Article
Published: Taylor & Francis 2017
Subjects:
Online Access:https://eprints.nottingham.ac.uk/46952/
_version_ 1848797435448524800
author Trapani, Lorenzo
author_facet Trapani, Lorenzo
author_sort Trapani, Lorenzo
building Nottingham Research Data Repository
collection Online Access
description This paper proposes a procedure to estimate the number of common factors k in a static approximate factor model. The building block of the analysis is the fact that the first k eigenvalues of the covariance matrix of the data diverge, whilst the others stay bounded. On the grounds of this, we propose a test for the null that the i-th eigenvalue diverges, using a randomised test statistic based directly on the estimated eigenvalue. The test only requires minimal assumptions on the data, and no assumptions are required on factors, loadings or idiosyncratic errors. The randomised tests are then employed in a sequential procedure to determine k.
first_indexed 2025-11-14T20:03:50Z
format Article
id nottingham-46952
institution University of Nottingham Malaysia Campus
institution_category Local University
last_indexed 2025-11-14T20:03:50Z
publishDate 2017
publisher Taylor & Francis
recordtype eprints
repository_type Digital Repository
spelling nottingham-469522020-05-04T18:51:59Z https://eprints.nottingham.ac.uk/46952/ A randomised sequential procedure to determine the number of factors Trapani, Lorenzo This paper proposes a procedure to estimate the number of common factors k in a static approximate factor model. The building block of the analysis is the fact that the first k eigenvalues of the covariance matrix of the data diverge, whilst the others stay bounded. On the grounds of this, we propose a test for the null that the i-th eigenvalue diverges, using a randomised test statistic based directly on the estimated eigenvalue. The test only requires minimal assumptions on the data, and no assumptions are required on factors, loadings or idiosyncratic errors. The randomised tests are then employed in a sequential procedure to determine k. Taylor & Francis 2017-06-26 Article PeerReviewed Trapani, Lorenzo (2017) A randomised sequential procedure to determine the number of factors. Journal of the American Statistical Association . ISSN 1537-274X Approximate factor models Randomised tests Number of factors https://doi.org/10.1080/01621459.2017.1328359 doi:10.1080/01621459.2017.1328359 doi:10.1080/01621459.2017.1328359
spellingShingle Approximate factor models
Randomised tests
Number of factors
Trapani, Lorenzo
A randomised sequential procedure to determine the number of factors
title A randomised sequential procedure to determine the number of factors
title_full A randomised sequential procedure to determine the number of factors
title_fullStr A randomised sequential procedure to determine the number of factors
title_full_unstemmed A randomised sequential procedure to determine the number of factors
title_short A randomised sequential procedure to determine the number of factors
title_sort randomised sequential procedure to determine the number of factors
topic Approximate factor models
Randomised tests
Number of factors
url https://eprints.nottingham.ac.uk/46952/
https://eprints.nottingham.ac.uk/46952/
https://eprints.nottingham.ac.uk/46952/