Rank reduction for modeling stationary signals
Louis L. Scharf, D.W. Tufts · IEEE Transactions on Acoustics Speech and Signal Processing · 1987
Rank reduction is developed as a general principle for trading off model bias and model variance in the analysis and synthesis of signals. The principle is applied to three basic problems: stationary time series modeling, stationary time series whitening, and vector quantization. Each problem brings its own surprises and insights.