Double Exponential-Smoothing Neural Network for Foreign Exchange Rate Forecasting

Financial time series analysis is an important research area that can predict various economic indicators such as the foreign currency exchange rate. In this paper, a deep-learning-based model is proposed to forecast the foreign exchange rate. Since the currency market is volatile and susceptible to ongoing social and political events, the proposed model incorporates event sentiments to accurately predict the exchange rate. Moreover, as the currency market is heavily dependent upon highly volatile factors such as gold and crude oil prices, we considered these sensitive factors for exchange rate forecasting. The validity of the model is tested over three currency exchange rates, which are Pak Rupee to US dollar (PKR/USD), British pound sterling to US dollar (GBP/USD), and Hong Kong Dollar to US dollar (HKD/USD). The study also shows the importance of incorporating investor sentiment of local and foreign macro-level events for accurate forecasting of the exchange rate. We processed approximately 5.9 million tweets to extract major events’ sentiment. The results show that this deep-learning-based model is a better predictor of foreign currency exchange rate in comparison with statistical techniques normally employed for prediction. The results present evidence that the exchange rate of all the three countries is more exposed to events happening in the US.

Download Full-text

Foreign exchange rate forecasting with a virtual-expert partial-consensus fuzzy-neural approach for semiconductor manufacturers in Taiwan

International Journal of Industrial and Systems Engineering ◽

10.1504/ijise.2013.050546 ◽

2013 ◽

Vol 13 (1) ◽

pp. 73 ◽

Cited By ~ 1

Author(s):

Toly Chen

Keyword(s):

Exchange Rate ◽

Foreign Exchange ◽

Foreign Exchange Rate ◽

Exchange Rate Forecasting ◽

Fuzzy Neural ◽

Virtual Expert ◽

Partial Consensus

Download Full-text

PEMODELAN METODE BROWN’S DOUBLE EXPONENTIAL SMOOTHING (B-DES) DAN BROWN’S WEIGHTED EXPONENTIAL MOVING AVERAGE (B-WEMA) MENGGUNAKAN OPTIMASI LEVENBERG-MARQUARDT PADA JUMLAH WISATAWAN DI JAWA TENGAH

Jurnal Gaussian ◽

10.14710/j.gauss.v9i3.27956 ◽

2020 ◽

Vol 9 (3) ◽

pp. 316-325

Author(s):

Dilla Retno Deswita ◽

Abdul Hoyyi ◽

Tatik Widiharih

Keyword(s):

Foreign Exchange ◽

Optimal Parameter ◽

Moving Average ◽

Exponential Smoothing ◽

Forecasting Accuracy ◽

Central Java ◽

Tourism Sector ◽

Double Exponential ◽

Levenberg Marquardt ◽

Double Exponential Smoothing

The tourism sector is one of the national development priority sectors because it contributes to foreign exchange earnings, the development of business areas, and the absorption of investment and labor. In 2018 the tourism sector will become the second largest foreign exchange earner after oil palm. Foreign exchange contributed by the tourism sector in 2018 was US $ 19.29 billion, an increase of 15.4%. The increase in contributions was driven by an increase in the number of foreign tourist arrivals by 12.58%, domestic tourists by 12.37%, and from investment. Therefore it is necessary to study the forecasting of the number of tourists after seeing the great potential generated from the tourism sector. The data forecast is data on the number of tourists in Central Java, both foreign and domestic data. Both data shows the tendency of an upward trend pattern. So that both data can be analyzed using B-DESmethods (Brown's Double Exponential Smoothing) and B-WEMA (Brown's Weighted Exponential Moving Average)that are optimized with LM (Levenberg-Marquardt). Both methods are able to analyze trend patterned data without assumptions making it easier in the analysis process. In addition, the two methods in previous studies were able to produce a small forecasting accuracy. The MAPE (Mean Absolute Percentage Error) value out sample is used to compare the forecasting results of the two methods. The results of the implementation of LM optimization on the data of the number of domestic tourists obtained the optimal parameter value of the B-DES method is 0.21944386 with MAPE out sample 16.26516% and B-WEMA method is 0.219441 with MAPE out sample 16.26515%. While the data on the number of foreign tourists obtained the optimal parameter value of the B-DES method was 0.26213368 with the MAPE out of the sample 23.61278% and the B-WEMA method was 0.26213367 with the MAPE out the sample 23.61278%. This means that both methods have a good level of forecasting accuracy in the data on the number of domestic tourists and an adequate level of accuracy in the data on the number of foreign tourists. Keywords : B-DES, B-WEMA, Levenberg-Marquardt, Tourists in Central Java

Download Full-text