scientific article; zbMATH DE number 1070406

From MaRDI portal
Publication:4356884





This paper presents an estimation algorithm for time-varying parameters of linear and nonlinear autoregressive models. First, the parameter space is discretized into \(N\) subspaces, each of which is assigned its center to represent the hidden Markov parameter value. Then, the Viterbi algorithm is applied to estimate the hidden Markov state. The result of a simulation study for a linear autoregressive model of order 2 is given to illustrate the proposed idea. The reviewer feels that the poor English may irritate the reader and the proposed idea does not work for much more complex systems because of the computational burden. For the entire collection see [Zbl 0868.00069].











This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4356884)