We develop statistical tools for time series analysis of high-dimensional multivariate datasets, when a few core series are of principal interest and there are many potential ancillary predictive variables. Themethodology, based on Vector Autoregressions (VAR), handles the case where unrestricted fitting is precluded by the large number of serie...
Creator:
McElroy, Tucker (U.S. Bureau of the Census)
Created:
2018-02-21
Contributed By:
University of Minnesota, Institute for Mathematics and its Applications.