查詢結果分析
來源資料
頁籤選單縮合
題名 | 時間序列模式選取準則之比較 |
---|---|
作者姓名(中文) | 虞國興; 劉治均; | 書刊名 | 農業工程學報 |
卷期 | 41:1 1995.03[民84.03] |
頁次 | 頁15-26 |
分類號 | 351.819 |
關鍵詞 | 時間序列模式; 準則; Time series model; Criterion; |
語文 | 中文(Chinese) |
中文摘要 | 時間序列模式之選取攸關預測能力之優劣,如何有效、正確地判定模式,實為一重要之研究課題。近幾十年來許多學者,諸儒Akaike (1969,1974,1979)、Schwarz (1978)、Bhansali及Downham (1977)、Haman及Quinn (1979)、Shibata (1980)等,分別提出各種不同模式選取準則,然而至今尚無完整之比較與研究。本研究採用CAT, FPE, FPE[9057], AIC, SBC, BIC1, BIC2, BIC3, BEC, HQ及S等11種準則,利用數千組合成資料,配合動差法與最大概似法推估參數,以其選取正確模式之精確度,探討各準則之性質及適用性。研究結果顯示,整體而言,以BIC1準則之選取能力最佳,而CAT為最差。 |
英文摘要 | The forecasting accuracy highly depends on the performance of model selection in time series modeling. Therefore, how to effectively and accurately identify model is one of the most important research topics in time series analysis. During the past few decades, several resea-chers, such as Akaike (1969, 1974, 1979) 、 Schwarz (1978) 、 Bhansali and Downham (1977) 、 Haman and Quinn (1979) and Shibata (1980), proposed lots of model selection criteria. However, systematical studies and comparisons on these criteria have not been made. In the present study, 11 criteria, CAT, FPE[8e60], FPE[8e60], AIC, SBC, BIC1, BIC2, BIC3, BEC, HQ and S are used. The aptness and characteristics of these criteria are investigated by the study of synthetic data based on the accuracy of selection of ture model. Both of the method of moments and maximum likelihood method are used to estimate the parameters. The results indicate that BICI is the best for the accuracy, while CAT is the worst. |
本系統之摘要資訊系依該期刊論文摘要之資訊為主。