A comparison of multivariate forecasting procedures for economic time series uri icon

abstract

  • In the recent decade several multivariate time-series methods have become available for forecasting. As with univariate methods, choices must be made as to which methods to use in practice. This paper reports the results of out-of-sample forecasts for several well-known procedures. Three interesting sets of data are used, and the forecasts are made over a five-year period. The data and model specifications are available upon request, so that the statistics presented in this paper can be used as a basis of comparison for future research. 1985 Elsevier Science Publishers B.V. (North-Holland) All rights reserved.

published proceedings

  • International Journal of Forecasting

author list (cited authors)

  • Kling, J. L., & Bessler, D. A.

citation count

  • 52

complete list of authors

  • Kling, John L||Bessler, David A

publication date

  • January 1985