Another look into the factor model black box: factor interpretation and structural (in)stability - HAL-SHS - Sciences de l'Homme et de la Société Accéder directement au contenu
Pré-Publication, Document De Travail Année : 2020

Another look into the factor model black box: factor interpretation and structural (in)stability

Résumé

This paper considers two families of methods allowing to get interpretable factors without imposing their interpretation a priori: factor rotations and sparse PCA. Monte Carlo simulations show their performance in recovering the correct factor structure. In an empirical application with a large U.S. macroeconomic dataset, they recover the same factor structure, offering a clear economic interpretation. This factor representation seems more natural and informative than competitors, offering new lens to disentangle the driving forces within a factor model. In particular, the structural instability appears to be more limited, and of different nature than what was previously found.
Fichier principal
Vignette du fichier
wp_1944_2.pdf (4.77 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

halshs-02235543 , version 1 (01-08-2019)
halshs-02235543 , version 2 (05-02-2020)
halshs-02235543 , version 3 (31-08-2021)

Identifiants

  • HAL Id : halshs-02235543 , version 2

Citer

Thomas Despois, Catherine Doz. Another look into the factor model black box: factor interpretation and structural (in)stability. 2020. ⟨halshs-02235543v2⟩
762 Consultations
1639 Téléchargements

Partager

Gmail Facebook X LinkedIn More