學術產出-Theses

Article View/Open

Publication Export

Google ScholarTM

政大圖書館

Citation Infomation

  • No doi shows Citation Infomation
題名 臺灣疫情期間的通貨膨脹率預測 - 分量因子模型的應用
Taiwan Inflation Rate Forecast During Covid-19 Pandemic - An Application of Quantile Factor Model
作者 郭彥伶
Kuo, Yen-Ling
貢獻者 徐士勛
Hsu, Shih-Hsun
郭彥伶
Kuo, Yen-Ling
關鍵詞 通貨膨脹率預測
擴散指標分析法
主成分分析法
分量因子分析法
Inflation forecast
Diffusion index analysis
Principal component analysis
Quantile factor analysis
日期 2023
上傳時間 1-Mar-2024 13:51:44 (UTC+8)
摘要 本文使用Chen et al. (2021)提出的分量因子分析法 (quantile factor analysis),從台灣經濟新報資料庫(TEJ)蒐集89個與消費者物價直接或間接相關的時間序列資料,從中萃取分量因子 (quantile factor),進行疫情期間(2020年到2022年)臺灣的通貨膨脹率預測。我們比較AR模型 (Autoregression model, AR)、共同因子模型 (factor model) 以及分量因子模型 (quantile factor model) 在疫情期間的預測績效,發現無論是通膨走升或是回落時期,分量因子模型的預測能力皆優於傳統的AR模型及共同因子模型。此外,我們計算各模型的預測分數 (predictive score) ,發現共同因子模型以及分量因子模型的預測分數普遍高於AR模型,顯示因子模型可以提供更精確的通膨預測。 另外,若將89個變數依照其特性區分為物價相關變數、實質面變數與金融面變數後,再分別估計各類別的因子,我們發現能進一步提升分量因子模型的預測績效。此外,若進一步計算分量因子在疫情期間的預測貢獻度,我們發現分量因子平均而言能提升約7%的模型預測能力,顯示分量因子在疫情期間可以良好的捕捉我國通膨的未來趨勢,改善傳統AR模型的預測能力。
This paper aims at forecasting Taiwan Inflation rate during the pandemic period (2020 to 2022) using quantile factor model proposed by Chen.et al (2021). We collected 89 time series data related to consumer prices from Taiwan Economic Journal Database (TEJ) and extracted quantile factors to make predictions. We compared forecast performances of AR model, common factor model and quantile factor model, and found that the predictive power of quantile factor model was better than that of AR model and common factor model, whenever the inflation rates was rising or falling. In addition, we found that the predictive scores of common factor model and quantile factor model were generally higher than that of AR model, showing that factor model can provide more accurate inflation forecasts. Furthermore, we found that if the 89 variables are divided into price-related variables, real variables and financial variables, the predictive performance of quantile factor model can be further improved. Next, we calculated forecast contribution of quantile factors during pandemic, and found that quantile factors can improve forecasting power by about 7% on average, showing that quantile factors can well capture the future trend of inflation during the pandemic in Taiwan.
參考文獻 吳若瑋 (2015),「通貨膨脹率之預測」,《經濟論文》, 43(2), 253­285 。 侯德潛與徐千婷 (2002),「我國通貨膨脹預測模型之建立」,《中央銀行季刊》, 24(3), 9­40 。 徐士勛、黃裕烈與徐之強 (2018),「台灣基本通膨估值 (UIG) 之建構與分 析」,《中央銀行季刊》, 41(3), 29­58 。 黃朝熙 (2007),「台灣通貨膨脹預測」,《中央銀行季刊》, 29(1), 5­30 。 陳佩玗 (2013),「台灣地區短期通貨膨脹率之預測」,《中央銀行季刊》, 35(1), 63­90 。 葉盛與田慧琦 (2004),「台灣的物價情勢: 影響因素探析與計量實證模型應 用」,《中央銀行季刊》, 26(4), 69­116 。 楊麗芬與許玉雪 (2005),「臺灣地區消費者物價指數-單變量與多變量時間數列 模式之比較分析」,《中國統計學報》, 43(3), 281­311 。 Adrian, T., N. Boyarchenko, and D. Giannone (2019), “Vulnerable Growth,” American Economic Review, 109(4), 1263­1289. Amengual, D., and E. Sentana (2020), “Is a Normal Coplua the Right Copula?,” Journal of Business and Economics Statistics, 38(2), 350­366. Atkeson, A. , and L. E. Ohanian (2001), “Are Phillips Curves Useful for Forecasting Inflation?,” Federal Reserve Bank of Minneapolis Quarterly Review, 25(1), 2–11. Bai, J. , and S. Ng (2002), “Determining the Number of Factors in Approximate Factor Models,” Econometrica, 70(1), 191­221. Bernanke, B. S., J. Boivin, and P. Eliasz (2005), “Measuring the Effects of Monetary Policy: A Factor­Augmented Vector Autoregressive (FAVAR) Approach,” The Quarterly Journal of Economics, 120(1), 387–422. Chen, L., J. J. Dolado, and J. Gonzalo (2021), “Quantile Factor Models,” Econmetrica, 89(2), 875­910. Connor, G., and R. A. Korajczyk (1986), “Performance Measurement With the Arbitrage Pricing Theory,” Journal of Financial Economics, 15, 373­394. Connor, G., and R. A. Korajczyk (1993), “A Test for the Number of Factors in the Approximate Factor Model,” Journal of Financial Economics, 48(4), 1263­ 1291. Granger, C. W. J. , and P. Newbold (1974), “Spurious Regression in Econometrics,” Journal of Econometrics, 2, 111­120. Meyler, A., G. Kenny, and T. Quinn (1998), “Forecasting Irish Inflation Using ARIMA Models,” Central Bank and Financial Services Authority of Ireland Technical Paper Series, 1998(3), 1­48. Mishkin, F. S. (1990), “What Does the Term Structure Tell Us About Future Inflation?,” Journal of Monetary Economics, 25, 77­95. Stock, J. H., and M. W. Watson (1999), “Forecasting Inflation,” NBER Working Paper, No. 7023. Stock, J. H., and M. W. Watson (2002), “Macroeconomics Forecasting Using Diffusion Indexes,” American Statistical Association Journal of Business and Economic Statistics, 20(2), 147­162. Stockton, D. J. , and J. E. Glassman (1987), “An Evaluation of the Forecast Performance of Alternative Models of Inflation,” The Review of Economics and Statistics, 69(1), 108-­117.
描述 碩士
國立政治大學
經濟學系
110258015
資料來源 http://thesis.lib.nccu.edu.tw/record/#G0110258015
資料類型 thesis
dc.contributor.advisor 徐士勛zh_TW
dc.contributor.advisor Hsu, Shih-Hsunen_US
dc.contributor.author (Authors) 郭彥伶zh_TW
dc.contributor.author (Authors) Kuo, Yen-Lingen_US
dc.creator (作者) 郭彥伶zh_TW
dc.creator (作者) Kuo, Yen-Lingen_US
dc.date (日期) 2023en_US
dc.date.accessioned 1-Mar-2024 13:51:44 (UTC+8)-
dc.date.available 1-Mar-2024 13:51:44 (UTC+8)-
dc.date.issued (上傳時間) 1-Mar-2024 13:51:44 (UTC+8)-
dc.identifier (Other Identifiers) G0110258015en_US
dc.identifier.uri (URI) https://nccur.lib.nccu.edu.tw/handle/140.119/150201-
dc.description (描述) 碩士zh_TW
dc.description (描述) 國立政治大學zh_TW
dc.description (描述) 經濟學系zh_TW
dc.description (描述) 110258015zh_TW
dc.description.abstract (摘要) 本文使用Chen et al. (2021)提出的分量因子分析法 (quantile factor analysis),從台灣經濟新報資料庫(TEJ)蒐集89個與消費者物價直接或間接相關的時間序列資料,從中萃取分量因子 (quantile factor),進行疫情期間(2020年到2022年)臺灣的通貨膨脹率預測。我們比較AR模型 (Autoregression model, AR)、共同因子模型 (factor model) 以及分量因子模型 (quantile factor model) 在疫情期間的預測績效,發現無論是通膨走升或是回落時期,分量因子模型的預測能力皆優於傳統的AR模型及共同因子模型。此外,我們計算各模型的預測分數 (predictive score) ,發現共同因子模型以及分量因子模型的預測分數普遍高於AR模型,顯示因子模型可以提供更精確的通膨預測。 另外,若將89個變數依照其特性區分為物價相關變數、實質面變數與金融面變數後,再分別估計各類別的因子,我們發現能進一步提升分量因子模型的預測績效。此外,若進一步計算分量因子在疫情期間的預測貢獻度,我們發現分量因子平均而言能提升約7%的模型預測能力,顯示分量因子在疫情期間可以良好的捕捉我國通膨的未來趨勢,改善傳統AR模型的預測能力。zh_TW
dc.description.abstract (摘要) This paper aims at forecasting Taiwan Inflation rate during the pandemic period (2020 to 2022) using quantile factor model proposed by Chen.et al (2021). We collected 89 time series data related to consumer prices from Taiwan Economic Journal Database (TEJ) and extracted quantile factors to make predictions. We compared forecast performances of AR model, common factor model and quantile factor model, and found that the predictive power of quantile factor model was better than that of AR model and common factor model, whenever the inflation rates was rising or falling. In addition, we found that the predictive scores of common factor model and quantile factor model were generally higher than that of AR model, showing that factor model can provide more accurate inflation forecasts. Furthermore, we found that if the 89 variables are divided into price-related variables, real variables and financial variables, the predictive performance of quantile factor model can be further improved. Next, we calculated forecast contribution of quantile factors during pandemic, and found that quantile factors can improve forecasting power by about 7% on average, showing that quantile factors can well capture the future trend of inflation during the pandemic in Taiwan.en_US
dc.description.tableofcontents 1 緒論 1 1.1 研究動機與目的 1 1.2 研究架構 3 2 文獻回顧 4 2.1 通膨預測方法文獻回顧 4 2.2 臺灣通膨預測文獻回顧 7 3 研究方法與模型 10 3.1 第一階段: 共同因子萃取 10 3.1.1 PCA 主成分分析法 11 3.1.2 QFA 分量因子分析法 13 3.2 第二階段: 建立線性預測模型 15 3.2.1 AR: 自我迴歸模型 16 3.2.2 AR+PC: 共同因子模型 16 3.2.3 AR+PC+QF: 分量因子模型 17 3.3 評估預測模型準則 19 3.3.1 MSE 均方差 19 3.3.2 RMSE 相對均方差 19 3.3.3 預測分數 20 3.3.4 預測貢獻度分析 21 I 4 資料 23 4.1 資料來源與說明 23 4.2 資料處理方式 24 4.2.1 季節性調整 24 4.2.2 資料轉換方法與 ADF 檢定 24 4.2.3 標準化轉換 25 4.3 研究標的 25 5 實證結果 27 5.1 因子個數估計及因子相關性分析 27 5.2 樣本外預測績效評估 30 5.3 預測分數 34 5.4 預測貢獻度分析 36 6 結論與建議 37 參考文獻 39 附錄 42 A 臺灣消費者物價指數直接與間接資料 42 B ADF 單根檢定結果 45zh_TW
dc.format.extent 1368083 bytes-
dc.format.mimetype application/pdf-
dc.source.uri (資料來源) http://thesis.lib.nccu.edu.tw/record/#G0110258015en_US
dc.subject (關鍵詞) 通貨膨脹率預測zh_TW
dc.subject (關鍵詞) 擴散指標分析法zh_TW
dc.subject (關鍵詞) 主成分分析法zh_TW
dc.subject (關鍵詞) 分量因子分析法zh_TW
dc.subject (關鍵詞) Inflation forecasten_US
dc.subject (關鍵詞) Diffusion index analysisen_US
dc.subject (關鍵詞) Principal component analysisen_US
dc.subject (關鍵詞) Quantile factor analysisen_US
dc.title (題名) 臺灣疫情期間的通貨膨脹率預測 - 分量因子模型的應用zh_TW
dc.title (題名) Taiwan Inflation Rate Forecast During Covid-19 Pandemic - An Application of Quantile Factor Modelen_US
dc.type (資料類型) thesisen_US
dc.relation.reference (參考文獻) 吳若瑋 (2015),「通貨膨脹率之預測」,《經濟論文》, 43(2), 253­285 。 侯德潛與徐千婷 (2002),「我國通貨膨脹預測模型之建立」,《中央銀行季刊》, 24(3), 9­40 。 徐士勛、黃裕烈與徐之強 (2018),「台灣基本通膨估值 (UIG) 之建構與分 析」,《中央銀行季刊》, 41(3), 29­58 。 黃朝熙 (2007),「台灣通貨膨脹預測」,《中央銀行季刊》, 29(1), 5­30 。 陳佩玗 (2013),「台灣地區短期通貨膨脹率之預測」,《中央銀行季刊》, 35(1), 63­90 。 葉盛與田慧琦 (2004),「台灣的物價情勢: 影響因素探析與計量實證模型應 用」,《中央銀行季刊》, 26(4), 69­116 。 楊麗芬與許玉雪 (2005),「臺灣地區消費者物價指數-單變量與多變量時間數列 模式之比較分析」,《中國統計學報》, 43(3), 281­311 。 Adrian, T., N. Boyarchenko, and D. Giannone (2019), “Vulnerable Growth,” American Economic Review, 109(4), 1263­1289. Amengual, D., and E. Sentana (2020), “Is a Normal Coplua the Right Copula?,” Journal of Business and Economics Statistics, 38(2), 350­366. Atkeson, A. , and L. E. Ohanian (2001), “Are Phillips Curves Useful for Forecasting Inflation?,” Federal Reserve Bank of Minneapolis Quarterly Review, 25(1), 2–11. Bai, J. , and S. Ng (2002), “Determining the Number of Factors in Approximate Factor Models,” Econometrica, 70(1), 191­221. Bernanke, B. S., J. Boivin, and P. Eliasz (2005), “Measuring the Effects of Monetary Policy: A Factor­Augmented Vector Autoregressive (FAVAR) Approach,” The Quarterly Journal of Economics, 120(1), 387–422. Chen, L., J. J. Dolado, and J. Gonzalo (2021), “Quantile Factor Models,” Econmetrica, 89(2), 875­910. Connor, G., and R. A. Korajczyk (1986), “Performance Measurement With the Arbitrage Pricing Theory,” Journal of Financial Economics, 15, 373­394. Connor, G., and R. A. Korajczyk (1993), “A Test for the Number of Factors in the Approximate Factor Model,” Journal of Financial Economics, 48(4), 1263­ 1291. Granger, C. W. J. , and P. Newbold (1974), “Spurious Regression in Econometrics,” Journal of Econometrics, 2, 111­120. Meyler, A., G. Kenny, and T. Quinn (1998), “Forecasting Irish Inflation Using ARIMA Models,” Central Bank and Financial Services Authority of Ireland Technical Paper Series, 1998(3), 1­48. Mishkin, F. S. (1990), “What Does the Term Structure Tell Us About Future Inflation?,” Journal of Monetary Economics, 25, 77­95. Stock, J. H., and M. W. Watson (1999), “Forecasting Inflation,” NBER Working Paper, No. 7023. Stock, J. H., and M. W. Watson (2002), “Macroeconomics Forecasting Using Diffusion Indexes,” American Statistical Association Journal of Business and Economic Statistics, 20(2), 147­162. Stockton, D. J. , and J. E. Glassman (1987), “An Evaluation of the Forecast Performance of Alternative Models of Inflation,” The Review of Economics and Statistics, 69(1), 108-­117.zh_TW