A panel of ex-ante forecasts of a single time series is modeled as a dynamic factor model, where the conditional expectation is the single unobserved factor. When applied to out-of-sample forecasting, this leads to combination forecasts that are based on methods other than OLS. These methods perform well in a Monte Carlo experiment. These methods are evaluated empirically in a panel of simulated real-time computer-generated univariate forecasts of U.S. macroeconomic time series.
About this article
Cite this article
Chan, Y., Stock, J. & Watson, M. A dynamic factor model framework for forecast combination. Span Econ Rev 1, 91–121 (1999). https://doi.org/10.1007/s101080050005