You are here

Determining the number of factors in approximate factor models with global and group-specific factors

Maximiano Pinheiro
Francisco Craveiro Dias
Publication Year 
JEL Code 
C32 - Time-Series Models
C52 - Model Evaluation and Testing
For an approximate factor model, in a static representation, with a common component comprising global factors and factors specific to groups of variables, the consistency of the principal components estimator is discussed. An extension of the well known Bai and Ng criteria is proposed for determining the number of global and group-specific factors. The consistency of the suggested criteria is established and the small sample properties are assessed through Monte Carlo simulations. As an empirical illustration, the proposed criteria is applied to estimate the number of global and country-specific macroeconomic factors for the major euro area countries.
Document link 
Published as