Title :
Intermittent Estimation for Gaussian Processes
Author :
Molnár-Sáska, Gábor ; Morvai, Gusztáv
Author_Institution :
Morgan Stanley Hungary Analytics, Ltd., Budapest, Hungary
fDate :
6/1/2010 12:00:00 AM
Abstract :
Let {Xn}n=0 ¿ be a stationary real-valued Gaussian time series. We estimate the conditional expectation E(Xn+1|X0, ...,Xn) from a growing number of observations X0,..., Xn in a pointwise consistent way along a sequence of stopping times.
Keywords :
Gaussian processes; information theory; time series; Gaussian process; conditional expectation; intermittent estimation; stationary real-valued Gaussian time series; stopping time; Gaussian distribution; Gaussian processes; H infinity control; Markov processes; Scholarships; Stochastic processes; Technological innovation; Conditional expectation; Gaussian process; estimation; stopping time;
Journal_Title :
Information Theory, IEEE Transactions on
DOI :
10.1109/TIT.2010.2046219
