Abstract | ||
---|---|---|
This paper examines the estimation of the order of an autoregressive model using the minimum description length principle. A closed form for an approximation of the parametric complexity of the autoregressive model class is derived by exploiting a relationship between coefficients and partial autocorrelations. The parametric complexity over the complete parameter space is found to diverge. A model selection criterion is subsequently derived by bounding the parameter space, and simulations suggest that it compares well against standard autoregressive order selection techniques in terms of correct order identification and prediction error. |
Year | DOI | Venue |
---|---|---|
2011 | 10.1109/TSP.2010.2091956 | IEEE Transactions on Signal Processing |
Keywords | Field | DocType |
standard autoregressive order selection,correct order identification,autoregressive model,parametric complexity,model selection criterion,complete parameter space,autoregressive model class,normalized maximum likelihood,minimum description length principle,closed form,parameter space,autoregressive process,computational modeling,correlation,maximum likelihood estimation,computer model,maximum likelihood estimate,minimum description length,model selection,stochastic processes,data models,stochastic process,gaussian processes | Applied mathematics,Autoregressive model,Mathematical optimization,Nonlinear autoregressive exogenous model,Minimum description length,Model selection,Autoregressive integrated moving average,Parametric statistics,Statistics,STAR model,Mathematics,Autocorrelation | Journal |
Volume | Issue | ISSN |
59 | 2 | 1053-587X |
Citations | PageRank | References |
4 | 0.39 | 8 |
Authors | ||
2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Daniel F. Schmidt | 1 | 51 | 10.68 |
Enes Makalic | 2 | 55 | 11.54 |