Application of ARIMA Model in Forecasting Exchange Rate: Evidence from Bangladesh
Keywords:ARIMA, Exchange Rate, Exchange Rate Forecasting, Time Series Model
This paper attempts to apply the ARIMA time series model to forecast the exchange rate of seven currencies (United States Dollar, Euro, Pound sterling, Australian Dollar, Japanese Yen, Canadian Dollar and Swedish Krona) in terms of Bangladeshi Taka (BDT) and to investigate the accuracy of the model by comparing the forecasted rates with the actual exchange rates. It considered daily currency exchange rates (244 selling price) of seven currencies for twelve months from January 2018 to December 2018 to forecast the subsequent one month (25 selling rate) in January 2019 rate. The Durbin-Watson test result shows an autocorrelation in the daily foreign currency exchange rate with the previous rate. The Augmented Dickey-Fuller test result shows data have unit roots and non-stationary. But the 1st differencing becomes data stationary to apply d equal to 1 in ARIMA model. Also, autocorrelation function considers MA(0) and partial autocorrelation function considers AR(1) for the ARIMA model. So, ARIMA (1,1,0) models are selected based on Ljung-Box test, root mean square error, mean absolute percent error, mean absolute error and R-square values. By using the above ARIMA models, forecasted foreign currency exchange rates next one month calculated and compared with the respective actual rates, which validate with Chi-Square test, mean absolute percent error, mean square error, root mean square error values of Goodness fit test. The result shows that predicted foreign currency exchange rates follow ARIMA (1,1,0) model, which may be applied to forecast the foreign currency exchange rates in Bangladesh.
Ahmed, F., & Keya, J. A. (2019). The Time Series Analysis forPredicting the Exchange Rate of USD to BDT. International Journal of Academic Research in Business, Arts and Science, 1(2). 282-294.
Alam, M. J. (2012). Forecasting the BDT/USD Exchange Rate using Autoregressive Model. Global Journal of Management and Business Research. 12(19).
Andreou, A. S., Georgpoulus, E. F., & Likothanassis, S. D. (2002).Exchange-Rates Forecasting: A hybrid Algorithm Based onGenetically Optimized Adaptative Neural Networks. Computational Economics, 20, 191-210.
Chinn, M., & Meese, D. (1995). Banking on currency forecasts: how predictable is the change in money? Journal of InternationalEconomics, 38, 161-178.
Chowdhury, T. U., & Islam, M. S. (2021). ARIMA Time SeriesAnalysis in Forecasting Daily Stock Price of Chittagong StockExchange (CSE). International Journal of Research and Innovationin Social Science, 5(6), 214-233.
Dunis, D. L., & Chen, Y. X. (2006). Alternative volatility models for risk management and trading: Application to the EUR/USD and USD/JPY rates. Derivatives Use, Trading & Regulation, 11(2), 126-156. DOI: https://doi.org/10.1057/palgrave.dutr.1840013
Goldberg, M. D., & Frydman, R. (1996). Empirical Exchange RateModels and Shifts in the Cointegrating Vector. Structural Changeand Economics Dynamics, 7. 55-78.
Hwang, J.K. (2001). Dynamic Forecasting of Monetary ExchangeRate Models: Evidence from Cointegration. International Advancein Economic Research, 7. 51-64.
Khashei, M., & Mahdavi Sharif, B. (2020). A Kalman filter-based hybridization model of statistical and intelligent approaches for exchange rate forecasting. Journal of Modelling in Management,ahead of print, ahead of print. DOI: https://doi.org/10.1108/JM2-12-2019-0277
Kilian, L., & Taylor, M. P. (2001). Why is it so difficult to beat therandom walk forecast of exchange rates? Working Papers, ResearchSeminar in International Economics, University of Michigan, Nr.464.
MacDonald, R., & Marsh, I. W. (1994). Combining exchange rateforecasts: What is the optimal consensus measure? Journal ofForecasting, 13, 313-333.
Mark, N. (1995). Exchange rates and fundamentals: evidence onlong-horizon predictability. American Economic Review, 201-218.
Marsh, I. W., & Power, D. M. (1996). A note on the performance offoreign exchange forecasters in a portfolio framework. Journal ofBanking Finance, 20, 605-613.
Matroushi. S. (2011). Hybrid computational intelligence systemsbased on statistical and neural networks methods for time seriesforecasting: The case of the gold price [Master Thesis, LincolnUniversity]. Retrieved from https://researcharchive.lincoln.ac.nz/handle/10182/3986.
Meese, R., & Rogoff, K. (1983).The out-of-sample failure ofempirical exchange rates: sampling error or misspecification? inFrenkel, J Exchange Rates and International Macroeconomics, 67-105, University of Chicago Press.
Mucaj, R., & Sinaj, V. (2017). Exchange rate forecasting usingARIMA, NAR, and ARIMA-ANN Hybrid model. Journal ofMultidisciplinary Engineering Science and Technology, 4(10), 8581-8586. Retrieved from http://www.jmest.org/wp-content/uploads/ JMESTN42352478.pdf
Wang, S., Tang. Z., & Chai. B. (2016). Exchange rate prediction model analysis based on improved artificial neural network algorithm. 2016 International Conference on Communication andElectronics Systems (ICCES), Coimbatore, India, 21-26, 1-5. Retrieved from https://doi.org/10.1109/CESYS.2016.7889912.