Ignoring cross-correlated idiosyncratic components when extracting factors in dynamic factor models

DOI10.1016/J.ECONLET.2023.111246MaRDI QIDQ6093785FDOQ6093785

Authors: Diego E. Fresoli, Pilar Poncela, Esther Ruiz 

Publication date: 12 September 2023

Published in: Economics Letters (Search for Journal in Brave)

zbMATH Keywords

Kalman filterprincipal componentsEM algorithmstate-space model

Mathematics Subject Classification ID

Statistics (62-XX)

Cites Work

• Are more data always better for factor analysis?
• Inferential Theory for Factor Models of Large Dimensions
• A two-step estimator for large approximate dynamic factor models based on Kalman filtering
• Principal components estimation and identification of static factors
• Confidence Intervals for Diffusion Index Forecasts and Inference for Factor-Augmented Regressions
• Statistical analysis of factor models of high dimension
• Bootstrap prediction mean squared errors of unobserved states based on the Kalman filter with estimated parameters
• Computing the mean square error of unobserved components extracted by misspecified time series models
• Efficient estimation of factor models
• Factor Extraction in Dynamic Factor Models: Kalman Filter Versus Principal Components

Cited In (1)

• Detecting irrelevant variables in possible proxies for the latent factors in macroeconomics and finance

This page was built for publication: Ignoring cross-correlated idiosyncratic components when extracting factors in dynamic factor models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6093785)