Download Now Free registration required
The estimation of dynamic factor models for large cross-sections poses a challenge in a real time environment. As macroeconomic data become available with different delays, unbalanced panel data sets with missing values at the end of the sample period (the so called "Jagged edge") have to be handled when estimating the factor model. In this paper, the author proposes an EM algorithm which copes with such data sets, accounts for autoregressive common factors and allows for serial correlation in the idiosyncratic components.
- Format: PDF
- Size: 451.64 KB