Early warning systems via machine learning: a study of currency crises
DOI:
https://doi.org/10.11606/1980-5330/ea189768Palavras-chave:
crises cambiais, previsão, machine learning, aprendizado de máquinaResumo
Sistemas de Alerta Antecipado (EWS) para crises cambiais é um tópico essencial em macroeconomia. renovado por métodos de aprendizado de máquina. A maioria dos trabalhos publicados tem métricas de precisão excessivamente otimistas causadas pela desconsideração da autocorrelação ou atrasos na publicação de dados. Nossa contribuição é construir um Sistema de Alerta Antecipado baseado em um conjunto de modelos de aprendizado de máquina apropriados para dados de séries temporais. Usando dados de 25 países entre 1995 e 2020, nossas descobertas são mais modestas do que trabalhos recentes, mas destacam a utilidade e as limitações dos sistemas de alerta antecipado na prática.
Downloads
Referências
ABOLGHASEMi, Mahdi et al. Machine learning applications in time series hierarchical forecasting. arXiv preprint arXiv:1912.00370, 2019.
ALAMINOS, David et al. Currency crises prediction using deep neural decision trees. Applied Sciences, v. 9, n. 23, p. 5227, 2019.
AMPOMAH, Ernest Kwame; QIN, Zhiguang; NYAME, Gabriel. Evaluation of Tree-Based Ensemble Machine Learning Models in Predicting Stock Price Direction of Movement. Information, v. 11, n. 6, p. 332, 2020.
BAŃBURA, Marta; RÜNSTLER, Gerhard. A look into the factor model black box: publication lags and the role of hard and soft data in forecasting GDP. International Journal of Forecasting, v. 27, n. 2, p. 333–346, 2011.
BERGMEIR, Christoph; HYNDMAN, Rob J.; KOO, Bonsoo. A note on the validity of cross-validation for evaluating autoregressive time series prediction. Computational Statistics & Data Analysis, v. 120, p. 70–83, 2018.
BEUTEL, Johannes; LIST, Sophia; VON SCHWEINITZ, Gregor. An evaluation of early warning models for systemic banking crises: Does machine learning improve predictions? Deutsche Bundesbank Discussion Paper No. 48/2018, 2018.
BOONMAN, Tjeerd M. et al. Early warning systems for currency crises with real time data. Open Economies Review, v. 30, n. 4, p. 813–835, 2019.
BUSSIERE, Matthieu; FRATZSCHER, Marcel. Towards a new early warning system of financial crises. Journal of International Money and Finance, v. 25, n. 6, p. 953–973, 2006.
CANDELON, Bertrand; DUMITRESCU, Elena-Ivona; HURLIN, Christophe. Currency crisis early warning systems: Why they should be dynamic. International Journal of Forecasting, v. 30, n. 4, p. 1016–1029, 2014.
CANDELON, Bertrand; DUMITRESCU, Elena-Ivona; HURLIN, Christophe. How to evaluate an early-warning system: Toward a unified statistical framework for assessing financial crises forecasting methods. IMF Economic Review, v. 60, n. 1, p. 75–113, 2012.
CERQUEIRA, Vitor; TORGO, Luis; MOZETIČ, Igor. Evaluating time series forecasting models: An empirical study on performance estimation methods. Machine Learning, v. 109, n. 11, p. 1997–2028, 2020.
CHONG, Terence T. L.; YAN, Isabel K. Forecasting currency crises with threshold models. International Economics, v. 156, p. 156–174, 2018.
CUARESMA, Jesús Crespo et al. Determinants of currency crises: A conflict of generations? Focus on European Economic Integration, n. 1, p. 126–141, 2008.
DU, Jiangze; YU, Runfang; LAI, Kin Keung. Identification and prediction of currency crisis: Markov Switching-Based approach. The Singapore Economic Review, v. 65, n. 06, p. 1667–1698, 2020.
EICHENGREEN, Barry; ROSE, Andrew K.; WYPLOSZ, Charles. Exchange market mayhem: the antecedents and aftermath of speculative attacks. Economic Policy, v. 10, n. 21, p. 249–312, 1995.
EMIN, Dogus; AYTAÇ, Ayşegül. Predicting currency crises: how do indicators differ according to crisis definition? Çankırı Karatekin University Journal of The Faculty of Economics and Administrative Sciences Y, v. 6, n. 2, p. 85–102, 2016.
FLOOD, Robert P.; GARBER, Peter M. Collapsing exchange-rate regimes: Some linear examples. Journal of International Economics, v. 17, n. 1-2, p. 1–13, 1984.
GÓMEZ, Víctor; MARAVALL, Agustín. Estimation, prediction, and interpolation for nonstationary series with the Kalman filter. Journal of the American Statistical Association, v. 89, n. 426, p. 611–624, 1994.
HASTIE, Trevor; TIBSHIRANI, Robert; FRIEDMAN, Jerome. The elements of statistical learning. New York: Springer, 2009.
HO, Tin Kam. The random subspace method for constructing decision forests. IEEE transactions on pattern analysis and machine intelligence, v. 20, n. 8, p. 832–844, 1998.
HOLOPAINEN, Markus; SARLIN, Peter. Toward robust early-warning models: A horse race, ensembles and model uncertainty. Quantitative Finance, v. 17, n. 12, p. 1933–1963, 2017.
KAMINSKY, Graciela; LIZONDO, Saul; REINHART, Carmen M. Leading indicators of currency crises. IMF Staff Papers, v. 45, n. 1, p. 1–48, 1998.
KINKYO, Takuji. A bi-annual forecasting model of currency crises. Applied Economics Letters, v. 27, n. 4, p. 255–261, 2020.
KRUGMAN, Paul. A model of balance-of-payments crises. Journal of Money, Credit and Banking, v. 11, n. 3, p. 311–325, 1979.
KRUGMAN, Paul. Crises: The Next Generation? In: Economic Policy in the International Economy: Essays in Honor of Assaf Razin. Organized by: E. Helpman and E. Sadka. Cambridge: Cambridge University Press, 2003. p. 15–32.
LOHRMANN, Christoph; LUUKKA, Pasi. Classification of intraday S&P500 returns with a Random Forest. International Journal of Forecasting, v. 35, n. 1, p. 390–407, 2019.
MALKIEL, Burton G.; FAMA, Eugene F. Efficient capital markets: A review of theory and empirical work. The Journal of Finance, v. 25, n. 2, p. 383–417, 1970.
MASSON, Paul R. Contagion: monsoonal effects, spillovers, and jumps between multiple equilibria. Washington, DC: International Monetary Fund, 1998.
MONTERO-MANSO, Pablo et al. FFORMA: Feature-based forecast model averaging. International Journal of Forecasting, v. 36, n. 1, p. 86–92, 2020.
NAKATANI, Ryota. Output costs of currency crises: shocks, policies and cycles. Munich: Munich Personal RePEc Archive, 2018. (MPRA Paper No. 83549).
OBSTFELD, Maurice. Models of currency crises with self-fulfilling features. European Economic Review, v. 40, n. 3-5, p. 1037–1047, 1996.
RAMLI, Nor Azuana; ISMAIL, Mohd Tahir; WOOI, Hooy Chee. Measuring the accuracy of currency crisis prediction with combined classifiers in designing early warning system. Machine Learning, v. 101, n. 1, p. 85–103, 2015.
SCHNAUBELT, Matthias. A comparison of machine learning model validation schemes for non-stationary time series data. Nürnberg: Friedrich-Alexander-Universität Erlangen-Nürnberg, 2019. (FAU Discussion Papers in Economics No. 11/2019).
SEVIM, Cuneyt et al. Developing an early warning system to predict currency crises. European Journal of Operational Research, v. 237, n. 3, p. 1095–1104, 2014.
SIAMI-NAMINI, Sima; NAMIN, Akbar Siami. Forecasting economics and financial time series: ARIMA vs. LSTM. arXiv preprint arXiv:1803.06386, 2018.
ZHANG, Cha; ma, Yunqian. Ensemble machine learning: methods and applications. New York: Springer, 2012.
Downloads
Publicado
Edição
Seção
Licença
Copyright (c) 2023 Economia Aplicada
Este trabalho está licenciado sob uma licença Creative Commons Attribution-NonCommercial 4.0 International License.
