Ribeiro, M.I. and Moura, J.M.F., 1991, LD 2 -ARMA identification algorithm," IEEE Transactions on signal processing, Vol. 39, pp.1822-1835. 26 LIST OF ILLUSTRATIONS AND TABLES Fig. 1 Modeling errors versus order. (a) N=500, f f = - - ( . , . , . ) 0 8 01 012  Home/Search   Document Not in Database   Summary   Related Articles

.... The AIC selects orders based on the concept of entropy [2] The approach proposed by Fuchs is based on the rank determination of estimated covariance matrices [8] Ribeiro and Moura determine the order by minimizing a functional d that measures the mismatch of the assumed model to the data 2 [17, 18]. This functional is evaluated from the estimated reflection coefficient sequence associated with the process. In [9] and [4] an instrumental variable based and a Wald statistic based approach are proposed. Although the principles of these criteria may differ from one another, a similarity does ....

....with the ARMA process, without any direct consideration of the properties of the process which are related to the model application. The percentage of times in which order selected is the same as the actual models is calculated in simulations to show the superiority of the proposed approaches [8, 17, 18, 4], even if the sample sizes are small. However, the estimated model with the same order as the actual model may not be the most appropriate when the samples are limited (see Section 4) As a matter of fact, each individual property associated with the process is only a reflection of model goodness ....

Ribeiro, M.I. and Moura, J.M.F., 1991, LD 2 -ARMA identification algorithm," IEEE Transactions on signal processing, Vol. 39, pp.1822-1835. 26 LIST OF ILLUSTRATIONS AND TABLES Fig. 1 Modeling errors versus order. (a) N=500, f f = - - ( . , . , . ) 0 8 01 012

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC