Title :
Maximum entropy modeling of periodically correlated processes
Author_Institution :
Dept. of Math., Stat. & Comput. Sci., Marquette Univ., Milwaukee, WI, USA
fDate :
11/1/1997 12:00:00 AM
Abstract :
When the covariance function of a periodically correlated process is known to a certain lag, we show it can be extrapolated in such a way that it maximizes the entropy. The process with the maximum entropy is a Gaussian periodic autoregressive process and is unique in distribution
Keywords :
Gaussian distribution; Gaussian processes; autoregressive processes; correlation theory; covariance analysis; extrapolation; information theory; maximum entropy methods; Gaussian periodic autoregressive process; covariance function; extrapolation; maximum entropy modeling; periodically correlated processes; Autoregressive processes; Entropy; Extrapolation; Gaussian processes; Geography; Hydrology; Probability density function; Signal processing; Spectral analysis; Statistics;
Journal_Title :
Information Theory, IEEE Transactions on
