Title :
The estimation of the model order in exponential families
Author_Institution :
Dept. of Electr. Eng., Technion, Haifa, Israel
fDate :
9/1/1989 12:00:00 AM
Abstract :
Estimators are sought that achieve high exponential rate of decrease in the underestimation probability while keeping the overestimation probability exponent at a certain prescribed level. It is assumed that a given integer is known to upper-bound the true order. Some examples are given of possible applications of the proposed method in specific order estimation and hypothesis testing problems
Keywords :
Markov processes; parameter estimation; probability; Markov chain; exponential families; hypothesis testing problems; model order estimation; overestimation probability exponent; underestimation probability; Automatic control; Bayesian methods; Error probability; Estimation error; Information analysis; Information theory; Parameter estimation; Random variables; Testing; Time series analysis;
Journal_Title :
Information Theory, IEEE Transactions on
