Adaptive order selection for autoregressive models

Please use this identifier to cite or link to this item: https://ir.lib.ncu.edu.tw/handle/987654321/105705

Title:	Adaptive order selection for autoregressive models
Authors:	陳春樹;Chen, Chun-Shu;Lee, Yun-Huan;Hsu, Hung-Wei
Contributors:	理學院統計研究所
Keywords:	Akaike information criterion;Bayesian analysis;Bayesian information criterion;Criteria;Decision making models;Degrees of freedom;Estimating techniques;Estimators;generalized degrees of freedom;Mathematical analysis;Mathematical models;Mathematical problems;mean squared prediction error;model selection;Order parameters;Parameter estimation;Perturbation methods;Regression analysis;Time series;time series data
Date:	2014-01-01
Issue Date:	2026-04-23 12:48:32 (UTC+8)
Publisher:	Taylor and Francis Ltd.;Abingdon: Taylor & Francis
Abstract:	摘要： Autoregressive model is a popular method for analysing the time dependent data, where selection of order parameter is imperative. Two commonly used selection criteria are the Akaike information criterion (AIC) and the Bayesian information criterion (BIC), which are known to suffer the potential problems regarding overfit and underfit, respectively. To our knowledge, there does not exist a criterion in the literature that can satisfactorily perform under various situations. Therefore, in this paper, we focus on forecasting the future values of an observed time series and propose an adaptive idea to combine the advantages of AIC and BIC but to mitigate their weaknesses based on the concept of generalized degrees of freedom. Instead of applying a fixed criterion to select the order parameter, we propose an approximately unbiased estimator of mean squared prediction errors based on a data perturbation technique for fairly comparing between AIC and BIC. Then use the selected criterion to determine the final order parameter. Some numerical experiments are performed to show the superiority of the proposed method and a real data set of the retail price index of China from 1952 to 2008 is also applied for illustration. 出版者： Abingdon: Taylor & Francis 出版日期： 2014-09-02 出處： Journal of statistical computation and simulation, 2014-09, Vol.84 (9), p.1963-1974 版權： 2013 Taylor & Francis 2013 版權： Copyright Taylor & Francis Ltd. 2014 識別號： ISSN: 0094-9655 識別號： EISSN: 1563-5163 識別號： DOI: 10.1080/00949655.2013.776559
Appears in Collections:	[Graduate Institute of Statistics] journal & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	14	View/Open

社群 sharing

Loading...