Original scientific paper
Distribution of random quadratic forms arising in singular-spectrum analysis
V. G. Moskvina
Abstract
We derive explicit expressions for the coefficients of certain quadratic
forms arising in the singular-spectrum analysis, a novel technique of
time series analysis. We also derive expressions for the first two moments of these quadratic forms under the assumption that the time series is a sum of a linear trend and white noise.
Keywords
time series analysis; random quadratic forms; singular-spectrum analysis; trajectory matrix; centring
Hrčak ID:
860
URI
Publication date:
20.12.2000.
Visits: 1.130 *