Determining the number of primitive shocks in factor models

  • Bai J
  • Ng S
  • 68


    Mendeley users who have this article in their library.
  • 161


    Citations of this article.


A widely held but untested assumption underlying macroeconomic analysis is that the number of shocks driving economic fluctuations, q, is small. In this paper, we associate q with the number of dynamic factors in a large panel of data. We propose a methodology to determine q without having to estimate the dynamic factors. Our analysis is based on the residuals of a VAR in r static factors, where the factors are themselves obtained by applying the method of principal components to a large panel of data. Our algebraic tests are based on a spectral decomposition of the covariance matrix of the residuals of a VAR. The tests are exact if the residuals were observable. Because of sampling variability fromhaving to estimate the VAR, the tests are accurate up to an error that vanishes asymptotically. We show how this error depends on the rate of convergence of the sample covariance to the population covariance of the true innovations. An important aspect of the present analysis is to make precise the relation between the dynamic factors and the static factors, which is a result of independent interest.

Author-supplied keywords

  • Common shocks
  • Dynamic factor model
  • Number of factors
  • Principal components analysis
  • Static factor

Get free article suggestions today

Mendeley saves you time finding and organizing research

Sign up here
Already have an account ?Sign in

Find this document


  • Jushan Bai

  • Serena Ng

Cite this document

Choose a citation style from the tabs below

Save time finding and organizing research with Mendeley

Sign up for free