重新審視Meese-Rogoff謎團—機器學習方法在匯率預測上之應用

、線上人數：40

、訪客IP：18.117.254.177

姓名	葉冠汝(Yeh Kuan Ju) 查詢紙本館藏	畢業系所	經濟學系
論文名稱	重新審視Meese-Rogoff謎團—機器學習方法在匯率預測上之應用
檔案	[Endnote RIS 格式] [Bibtex 格式] [相關文章] [文章引用] [完整記錄] [館藏目錄] [檢視] [下載] 本電子論文使用權限為同意立即開放。已達開放權限電子全文僅授權使用者為學術研究之目的，進行個人非營利性質之檢索、閱讀、列印。請遵守中華民國著作權法之相關規定，切勿任意重製、散佈、改作、轉貼、播送，以免觸法。
摘要(中)	本研究探討了美元兌各個主要國家的月頻率樣本外匯率變動率預測，並引入多種機器學習模型來重新檢驗Meese and Rogoff (1983)的結論–隨機漫步為短期匯率預測最好的模型是否依然正確。本文以月頻率資料為基礎，選取了多個國家和全球經濟變數，並使用了收縮模型、主成分模型、隨機森林（Random Forest）、深度神經網絡（DNN）等多種機器學習模型，對美元兌主要貨幣（包括日元、歐元、澳元、加元和英鎊）的匯率變動率進行樣本外預測，並與隨機漫步模型進行比較。實證結果顯示，在領先一期的預測中，機器學習模型普遍優於隨機漫步模型。然而隨著領先期數增加至兩期和三期，機器學習模型的預測能力下降，優勢逐漸減弱，這樣的結果部分支持了Meese and Rogoff的結論。透過變數選擇模型可以了解個別變數的重要性，大部分變數以與全球變數的交乘項的型態被挑選，顯示在預測匯率時考慮全球局勢的重要性。Clark-West檢定的結果也同樣顯示多數機器學習模型在領先一期的預測中顯著優於隨機漫步模型，但在更長期的預測中，機器學習模型的優勢逐漸減少。本研究展示了機器學習方法在經濟、金融時間序列預測中的潛力，未來可進一步探索更多變數和改進模型結構，以提升長期預測的準確性。
摘要(英)	This study revisits the conclusions of Meese and Rogoff (1983), which suggested that the random walk model is the best for short-term exchange rate forecasting. Using monthly data and various machine learning models—including shrinkage models, principal component models, random forests, and deep neural networks (DNN)—we forecast the US dollar exchange rates against major currencies (yen, euro, Australian dollar, Canadian dollar, and British pound) and compare them with the random walk model. Empirical results show that machine learning models outperform the random walk model for one-month-ahead forecasts. However, their predictive power decreases for two and three-month-ahead forecasts, partially supporting Meese and Rogoff′s conclusions. LASSO analysis reveals that most important variables are global interactions, highlighting the significance of global factors in exchange rate predictions. The Clark-West test also confirms that most machine learning models are significantly better than the random walk model for short-term forecasts, though their advantage diminishes over longer horizons. In summary, machine learning models show significant potential for short-term exchange rate forecasting, particularly for one-month-ahead predictions. However, their accuracy declines over longer periods, not consistently outperforming the random walk model. Future research should explore additional variables and improved model structures to enhance long-term forecasting accuracy.
關鍵字(中)	★ 匯率預測 ★ 機器學習 ★ 隨機漫步	關鍵字(英)	★ Machine learning models ★ Exchange rate forecasting ★ Random walk
論文目次	中文摘要. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . i 英文摘要. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ii 致謝. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii 目錄. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iv 圖目錄. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi 表目錄. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii 一、緒論. . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 二、文獻回顧. . . . . . . . . . . . . . . . . . . . . . . . . 3 三、研究方法. . . . . . . . . . . . . . . . . . . . . . . . . 5 3-1各國匯率資料. . . . . . . . . . . . . . . . . . . . . . 5 3-2變數選擇. . . . . . . . . . . . . . . . . . . . . . . . . 6 3-3模型設定. . . . . . . . . . . . . . . . . . . . . . . . . 10 3-4基準模型(BenchmarkModel) . . . . . . . . . . . . . . 11 3-5線性預測模型. . . . . . . . . . . . . . . . . . . . . . 12 3-5-1向前逐步迴歸ForwardStepwiseSelection . . . . . . . 12 3-5-2縮減模型. . . . . . . . . . . . . . . . . . . . . . . . . 13 3-5-3 AdaptiveLASSO . . . . . . . . . . . . . . . . . . . . . 16 3-5-4主成分模型. . . . . . . . . . . . . . . . . . . . . . . . 17 3-6非線性預測模型. . . . . . . . . . . . . . . . . . . . . 18 3-6-1深度神經網絡(DeepNeuralNetworks) . . . . . . . . . 18 iv 3-6-2隨機森林(RandomForest) . . . . . . . . . . . . . . . . 19 3-6-3 XGBoost . . . . . . . . . . . . . . . . . . . . . . . . . 20 3-7混和預測模型. . . . . . . . . . . . . . . . . . . . . . 21 3-7-1 LASSO+隨機森林. . . . . . . . . . . . . . . . . . . . 21 3-7-2深度神經網絡+Ridge . . . . . . . . . . . . . . . . . . 22 3-8模型評估. . . . . . . . . . . . . . . . . . . . . . . . . 22 四、實證分析. . . . . . . . . . . . . . . . . . . . . . . . . 25 4-1美元兌日元匯率變動率預測結果. . . . . . . . . . . . 25 4-2美元兌歐元匯率變動率預測結果. . . . . . . . . . . . 29 4-3美元兌加幣匯率變動率預測結果. . . . . . . . . . . . 32 4-4美元兌澳幣匯率變動率預測結果. . . . . . . . . . . . 35 4-5美元兌英鎊匯率變動率預測結果. . . . . . . . . . . . 38 五、結論. . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 參考文獻. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
參考文獻	Álvarez-Díaz, M.andÁlvarez, A.(2005). Geneticmulti-modelcompositeforecast for non-linear prediction of exchange rates. Empirical Economics, 30:643–663. Bacchetta, P., Van Wincoop, E., and Beutler, T. (2010). Can parameter instability explain the meese-rogoff puzzle? In NBER International Seminar on Macroe conomics, volume 6, pages 125–173. The University of Chicago Press Chicago, IL. Baker, S. R., Bloom, N., and Davis, S. J. (2016). Measuring economic policy uncertainty. The Quarterly Journal of Economics, 131(4):1593–1636. Breiman, L. (2001). Random forests. Machine learning, 45:5–32. Caldara, D. and Iacoviello, M. (2022). Measuring geopolitical risk. American Economic Review, 112(4):1194–1225. Campbell, J. Y. and Thompson, S. B. (2008). Predicting excess stock returns out of sample: Can anything beat the historical average? The Review of Financial Studies, 21(4):1509–1531. Chen, T. and Guestrin, C. (2016). Xgboost: A scalable tree boosting system. In Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining, pages 785–794. Chen, W., Xu, H., Jia, L., and Gao, Y. (2021). Machine learning model for bitcoin exchange rate prediction using economic and technology determinants. Interna tional Journal of Forecasting, 37(1):28–43. Clark, T. E. and West, K. D. (2007). Approximately normal tests for equal predic tive accuracy in nested models. Journal of Econometrics, 138(1):291–311. Eichenbaum, M. S., Johannsen, B. K., and Rebelo, S. T. (2021). Monetary pol icy and the predictability of nominal exchange rates. The Review of Economic Studies, 88(1):192–228. Engel, C. (1994). Can the markovswitching modelforecast exchange rates? Jour nal of International Economics, 36(1-2):151–165. Filippou, I., Rapach, D., Taylor, M. P., and Zhou, G. (2023). Out-of-sample ex change rate prediction: A machine learning perspective. Available at SSRN 3455713. Gu,S., Kelly, B., andXiu, D.(2020). Empiricalassetpricingviamachinelearning. The Review of Financial Studies, 33(5):2223–2273. Hastie, T., Tibshirani, R., Friedman, J. H., and Friedman, J. H. (2009). The ele ments of statistical learning: data mining, inference, and prediction, volume 2. Springer. Hoerl, A. E. and Kennard, R. W. (1970). Ridge regression: Biased estimation for nonorthogonal problems. Technometrics, 12(1):55–67. Kelly, B. and Xiu, D. (2023). Financial machine learning. Foundations and Trends® in Finance, 13(3-4):205–363. Maas, A.L., Hannun, A.Y., andNg,A.Y.(2013). Rectifiernonlinearities improve neural network acoustic models. In Proc. icml, volume 30, page 3. Atlanta, GA. Mark, N. C. (1995). Exchange rates and fundamentals: Evidence on long-horizon predictability. The American Economic Review, pages 201–218. Meese, R. A. and Rogoff, K. (1983). Empirical exchange rate models of the sev enties: Do they fit out of sample? Journal of International Economics, 14(1 2):3–24. Nagel, S. (2021). Machine learning in asset pricing, volume 1. Princeton Univer sity Press. Rapach, D. and Zhou, G. (2013). Forecasting stock returns. In Handbook of Economic Forecasting, volume 2, pages 328–383. Elsevier. Rapach, D. E. and Wohar, M. E. (2006). The out-of-sample forecasting perfor manceofnonlinear models ofreal exchange rate behavior. International Journal of Forecasting, 22(2):341–361. Rossi, B. (2013). Exchange rate predictability. Journal of Economic Literature, 51(4):1063–1119. Tibshirani, R. (1996). Regression shrinkage and selection via the lasso. Journal of the Royal Statistical Society Series B: Statistical Methodology, 58(1):267–288. Turhan, I., Hacihasanoglu, E., and Soytas, U. (2013). Oil prices and emerging market exchange rates. Emerging Markets Finance and Trade, 49(sup1):21–36. Yin, X. C., Li, X., Wang, M. H., Qin, M., and Shao, X. F. (2021). Do economic policy uncertainty and its components predict china’s housing returns? Pacific Basin Finance Journal, 68:101575. Zou, H. (2006). The adaptive lasso and its oracle properties. Journal of the Amer ican statistical association, 101(476):1418–1429. Zou, H. and Hastie, T. (2005). Regularization and variable selection via the elastic net. Journal of the Royal Statistical Society Series B: Statistical Methodology, 67(2):301–320.
指導教授	徐之強廖志興	審核日期	2024-7-23
推文	facebook plurk twitter funp google live udn HD myshare reddit netvibes friend youpush delicious baidu
網路書籤	Google bookmarks del.icio.us hemidemi myshare

博碩士論文 111429007 詳細資訊