Page 90 - Contributed Paper Session (CPS) - Volume 6

CPS1832 Nur Fazliana Rahim et al.

Fuzzy rule base method for forecasting time
series data
2
1,2
Nur Fazliana Rahim , Mahmod Othman , Rajalingtam Sokkalingam
2
1 Centre for Pre-University Studies, Universiti Malaysia Sarawak, 94300 Kota Samarahan,
Sarawak, Malaysia
2 Fundamental and Applied Sciences Department, Universiti Teknologi PETRONAS, 32610 Seri
Iskandar, Perak, Malaysia

Abstract
In generating fuzzy rule of forecasting, Weighted Subsethood-Based
Algorithm (WSBA) were used to develops Foreign Exchange Rate (FER)
forecasting method. In the use of Fuzzy Time Series to develop fuzzy rules,
Fuzzy Rule Based Systems (FRBS) concept was implemented. The intention of
the recommended method is to improve the efficiency of time series
forecasting which offer higher predicting accuracy. In order to validate this
method, 5 years’ data of FER and three currency pairs was used as testing data
sets, which are Malaysian Ringgit (MYR), Japanese Yen (JPY), South Korea Won
(KRW) and Singapore Dollar (SGP). The forecasting precision of this method
was compared with the prior methods. In this paper, the results proved that
this method can minimize forecasting error and so that increase the accuracy
of FER forecasting values. The outcomes of this paper could be used as
another choice of method to obtain the fuzzy rules to get a superior forecast
values of FER.

Keywords
Forecasting Foreign Exchange Rate; Fuzzy Time Series; Weighted Subsethood
Based Algorithm

1. Introduction
In dealing with Fuzzy Time Series (FTS) forecasting, exchange rate is found
to be a fascinating subject. Taken into consideration, exchange rate takes a
crucial part in worldwide trade, handling of business risk and the country
economic condition. (Korol, 2014). Several factors can affect forecasting the
exchange rate, while affecting currency ratings such as inflation, interest rate,
national debt, employment data, political stability and economic performance
and etc. (Patel et al., 2014). Currently a rare situation in currency market when
central bank intervenes and when doing so it is often successful. In 1997 when
the currency depreciates causing Malaysia facing a rough time due to inflation
(Leu et al., 2009). This is why forecasting the Foreign Exchange Rate (FER) is
vital and by using appropriate forecasting model that can validly forecast the

79 | I S I W S C 2 0 1 9