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。

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.

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.