eduzhai > Applied Sciences > Computer Science >

A Regularized Factor-augmented Vector Autoregressive Model

  • KanKan
  • (0) Download
  • 20210425
  • Save

... pages left unread,continue reading

Document pages: 73 pages

Abstract: We propose a regularized factor-augmented vector autoregressive (FAVAR) modelthat allows for sparsity in the factor loadings. In this framework, factors mayonly load on a subset of variables which simplifies the factor identificationand their economic interpretation. We identify the factors in a data-drivenmanner without imposing specific relations between the unobserved factors andthe underlying time series. Using our approach, the effects of structuralshocks can be investigated on economically meaningful factors and on allobserved time series included in the FAVAR model. We prove consistency for theestimators of the factor loadings, the covariance matrix of the idiosyncraticcomponent, the factors, as well as the autoregressive parameters in the dynamicmodel. In an empirical application, we investigate the effects of a monetarypolicy shock on a broad range of economically relevant variables. We identifythis shock using a joint identification of the factor model and the structuralinnovations in the VAR model. We find impulse response functions which are inline with economic rationale, both on the factor aggregates and observed timeseries level.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×