Publications-Theses

Article View/Open

Publication Export

Google ScholarTM

NCCU Library

Citation Infomation

Related Publications in TAIR

題名 應用雙季節指數平滑模型於來店人數預測之研究
Applications of double seasonal exponential smoothing to store traffic forecasting
作者 許展源
貢獻者 翁久幸
許展源
關鍵詞 指數平滑
狀態空間模型
雙季節
卡爾曼濾波
日期 2018
上傳時間 27-Jul-2018 11:30:26 (UTC+8)
摘要 Holt-Winters 法是一種可以同時考慮線性趨勢以及季節性的指數平滑法, 對於單一週期性時間序列資料有不錯的預測效果。Taylor [12] 提出的雙季節 指數平滑法比 Holt-Winters 法多了一個季節影響,適合用在具有兩種週期性 的資料。另外,指數平滑法雖然簡易好用,但是並無機率模型。Hyndman et al. [6] 將指數平滑法表示成具有單一誤差來源的狀態空間模型。有了狀態 空間模型表示式, 便能寫出概似函數,進行參數估計及區間估計, 而且可以 很自然地新增外生變數於此模型中。然而,對於單一誤差的狀態空間模型, 目前文獻上並無討論如何以類似卡爾曼濾波器的方式,來進行狀態的更新。
     本研究的主要貢獻有兩點。首先是關於指數平滑法在來店人數預測的比 較。本論文使用美國服飾業的來店人數資料,我們發現該筆資料具有雙季 節性,使用 Taylor [12] 的雙季節指數平滑法,在預測上明顯優於單季節指 數平滑法。第二點貢獻是針對單一誤差的狀態空間模型,目前文獻上的更 新預測步驟,仍是延續原本的指數平滑法,本論文試著提出類似卡爾曼濾 波器的迭代更新法,來進行狀態更新。有了這種更新方法後,處理外生變 數就容易許多
參考文獻 [1] S. Tom Au, Guang-Qin Ma, and Shu-Ngai Yeung. Automatic forecasting of double seasonal time series with applications on mobility network traffic prediction. 2011.
     
     [2] Robert Goodell Brown. Statistical forecasting for inventory control. McGraw Hill, New York, 1959.
     
     [3] Phillip G Gould, Anne B Koehler, J. Keith Ord, Ralph D Snyder, Rob J Hyndman, and Farshid Vahid-Araghi. Forecasting time series with multiple seasonal patterns. European Journal of Operational Research, 2008.
     
     [4] Charles C Holt. Forecasting seasonals and trends by exponentially weighted moving averages. Carnegie Institute of Technology, Graduate school of Industrial Administration, 1957.
     
     [5] Rob J Hyndman and Yeasmin Khandakar. Automatic time series forecasting: the forecast package for R. Journal of Statistical Software, 26(3):1–22, 2008.
     
     [6] Rob J. Hyndman, Anne B. Koehler, J. Keith Ord, and Ralph D. Snyder. Forecasting with Exponential Smoothing:The State Space Approach. Springer, 2008.
     
     [7] Mohamed A. Ismail, Alyaa R. Zahran, and Eman M. Abd El-Metaal. Forecasting hourly electricity demand in egypt using double seasonal autoregressive integrated moving average model. 2015.
     
     [8] R. E. Kalman. A new approach to linear filtering and prediction problems. Journal of Basic Engineering, 82:35–45, 1960.
     
     [9] R. E. Kalman and R. S. Bucy. New results in linear filtering and new results in linear filtering and prediction theory. Journal of Basic Engineering, 83:95–108, 1961.
     
     [10] Reinaldo Castro Souza, Mônica Barros, and Cristina Vidigal C. de Miranda. Short term load forecasting using double seasonal exponential smoothing and interventions to account for holidays and temperature effects. 2007.
     
     [11] Ivan Svetunkov. smooth: Forecasting Using Smoothing Functions, 2018.
     
     [12] J. W. Taylor. Short-term electricity demand forecasting using double seasonal exponential smoothing. The Journal of the Operational Research Society, 2013.
     
     [13] Greg Welch and Gary Bishop. An introduction to the kalman filter. Technical report, Department of Computer Science University of North Carolina at Chapel Hill, 2006.
     
     [14] Peter R Winters. Forecasting sales by exponentially weighted moving averages. Management Science, 1960.
     
     [15] 施佩吟. 指數平滑模型應用於來店人數預測之研究. 碩, 國立政治大學, 2015.
描述 碩士
國立政治大學
統計學系
105354023
資料來源 http://thesis.lib.nccu.edu.tw/record/#G0105354023
資料類型 thesis
dc.contributor.advisor 翁久幸zh_TW
dc.contributor.author (Authors) 許展源zh_TW
dc.creator (作者) 許展源zh_TW
dc.date (日期) 2018en_US
dc.date.accessioned 27-Jul-2018 11:30:26 (UTC+8)-
dc.date.available 27-Jul-2018 11:30:26 (UTC+8)-
dc.date.issued (上傳時間) 27-Jul-2018 11:30:26 (UTC+8)-
dc.identifier (Other Identifiers) G0105354023en_US
dc.identifier.uri (URI) http://nccur.lib.nccu.edu.tw/handle/140.119/118933-
dc.description (描述) 碩士zh_TW
dc.description (描述) 國立政治大學zh_TW
dc.description (描述) 統計學系zh_TW
dc.description (描述) 105354023zh_TW
dc.description.abstract (摘要) Holt-Winters 法是一種可以同時考慮線性趨勢以及季節性的指數平滑法, 對於單一週期性時間序列資料有不錯的預測效果。Taylor [12] 提出的雙季節 指數平滑法比 Holt-Winters 法多了一個季節影響,適合用在具有兩種週期性 的資料。另外,指數平滑法雖然簡易好用,但是並無機率模型。Hyndman et al. [6] 將指數平滑法表示成具有單一誤差來源的狀態空間模型。有了狀態 空間模型表示式, 便能寫出概似函數,進行參數估計及區間估計, 而且可以 很自然地新增外生變數於此模型中。然而,對於單一誤差的狀態空間模型, 目前文獻上並無討論如何以類似卡爾曼濾波器的方式,來進行狀態的更新。
     本研究的主要貢獻有兩點。首先是關於指數平滑法在來店人數預測的比 較。本論文使用美國服飾業的來店人數資料,我們發現該筆資料具有雙季 節性,使用 Taylor [12] 的雙季節指數平滑法,在預測上明顯優於單季節指 數平滑法。第二點貢獻是針對單一誤差的狀態空間模型,目前文獻上的更 新預測步驟,仍是延續原本的指數平滑法,本論文試著提出類似卡爾曼濾 波器的迭代更新法,來進行狀態更新。有了這種更新方法後,處理外生變 數就容易許多
zh_TW
dc.description.tableofcontents 中文摘要... i
     
     Abstract ... ii
     
     目錄... iii
     
     表目錄... iv
     
     圖目錄... v
     
     第一章 緒論 ... 1
     
     第二章 文獻回顧... 3
     
     第三章 研究方法... 5
     第一節 Holt-Winters 法... 5
     第二節 雙季節指數平滑法... 6
     第三節 卡爾曼濾波... 7
     第四節 創新狀態空間模型... 10
     
     第四章 創新狀態空間模型的估計 ... 14
     第一節 平滑係數選取... 14
     第二節 系統狀態更新... 15
     第三節 模擬... 20
     
     第五章 實證研究... 21
     第一節 資料介紹與分析步驟... 21
     第二節 模型分析... 23
     
     第六章 結論與建議... 29
     
     參考文獻... 31
zh_TW
dc.source.uri (資料來源) http://thesis.lib.nccu.edu.tw/record/#G0105354023en_US
dc.subject (關鍵詞) 指數平滑zh_TW
dc.subject (關鍵詞) 狀態空間模型zh_TW
dc.subject (關鍵詞) 雙季節zh_TW
dc.subject (關鍵詞) 卡爾曼濾波zh_TW
dc.title (題名) 應用雙季節指數平滑模型於來店人數預測之研究zh_TW
dc.title (題名) Applications of double seasonal exponential smoothing to store traffic forecastingen_US
dc.type (資料類型) thesisen_US
dc.relation.reference (參考文獻) [1] S. Tom Au, Guang-Qin Ma, and Shu-Ngai Yeung. Automatic forecasting of double seasonal time series with applications on mobility network traffic prediction. 2011.
     
     [2] Robert Goodell Brown. Statistical forecasting for inventory control. McGraw Hill, New York, 1959.
     
     [3] Phillip G Gould, Anne B Koehler, J. Keith Ord, Ralph D Snyder, Rob J Hyndman, and Farshid Vahid-Araghi. Forecasting time series with multiple seasonal patterns. European Journal of Operational Research, 2008.
     
     [4] Charles C Holt. Forecasting seasonals and trends by exponentially weighted moving averages. Carnegie Institute of Technology, Graduate school of Industrial Administration, 1957.
     
     [5] Rob J Hyndman and Yeasmin Khandakar. Automatic time series forecasting: the forecast package for R. Journal of Statistical Software, 26(3):1–22, 2008.
     
     [6] Rob J. Hyndman, Anne B. Koehler, J. Keith Ord, and Ralph D. Snyder. Forecasting with Exponential Smoothing:The State Space Approach. Springer, 2008.
     
     [7] Mohamed A. Ismail, Alyaa R. Zahran, and Eman M. Abd El-Metaal. Forecasting hourly electricity demand in egypt using double seasonal autoregressive integrated moving average model. 2015.
     
     [8] R. E. Kalman. A new approach to linear filtering and prediction problems. Journal of Basic Engineering, 82:35–45, 1960.
     
     [9] R. E. Kalman and R. S. Bucy. New results in linear filtering and new results in linear filtering and prediction theory. Journal of Basic Engineering, 83:95–108, 1961.
     
     [10] Reinaldo Castro Souza, Mônica Barros, and Cristina Vidigal C. de Miranda. Short term load forecasting using double seasonal exponential smoothing and interventions to account for holidays and temperature effects. 2007.
     
     [11] Ivan Svetunkov. smooth: Forecasting Using Smoothing Functions, 2018.
     
     [12] J. W. Taylor. Short-term electricity demand forecasting using double seasonal exponential smoothing. The Journal of the Operational Research Society, 2013.
     
     [13] Greg Welch and Gary Bishop. An introduction to the kalman filter. Technical report, Department of Computer Science University of North Carolina at Chapel Hill, 2006.
     
     [14] Peter R Winters. Forecasting sales by exponentially weighted moving averages. Management Science, 1960.
     
     [15] 施佩吟. 指數平滑模型應用於來店人數預測之研究. 碩, 國立政治大學, 2015.
zh_TW
dc.identifier.doi (DOI) 10.6814/THE.NCCU.STAT.008.2018.B03-