A Quasi–Maximum Likelihood Approach for Large, Approximate Dynamic Factor Models
Top Cited Papers
- 1 November 2012
- journal article
- Published by MIT Press in Review of Economics and Statistics
- Vol. 94 (4) , 1014-1024
Abstract
International audienceIs maximum likelihood suitable for factor models in large cross-sections of time series? We answer this question from both an asymptotic and an empirical perspective. We show that estimates of the common factors based on maximum likelihood are consistent for the size of the cross-section (n) and the sample size (T) going to in¯nity along any path and that maximum likelihood is viable for n large. The estimator is robust to misspecification of cross-sectional and time series correlation of the idiosyncratic components. In practice, the estimator can be easily implemented using the Kalman smoother and the EM algorithm as in traditional factor analysisKeywords
All Related Versions
This publication has 14 references indexed in Scilit:
- A two-step estimator for large approximate dynamic factor models based on Kalman filteringJournal of Econometrics, 2011
- Large Dimensional Factor AnalysisFoundations and Trends® in Econometrics, 2008
- Confidence Intervals for Diffusion Index Forecasts and Inference for Factor-Augmented RegressionsEconometrica, 2006
- The Generalized Dynamic Factor ModelJournal of the American Statistical Association, 2005
- International Business Cycles: World, Region, and Country-Specific FactorsAmerican Economic Review, 2003
- Inferential Theory for Factor Models of Large DimensionsEconometrica, 2003
- Determining the Number of Factors in Approximate Factor ModelsEconometrica, 2002
- Federal policies and local economies: Europe and the USEuropean Economic Review, 2001
- Funds, Factors, and Diversification in Arbitrage Pricing ModelsEconometrica, 1983
- Maximum Likelihood "Confirmatory" Factor Analysis of Economic Time SeriesInternational Economic Review, 1981