Page 95 - Contributed Paper Session (CPS) - Volume 6

CPS1832 Nur Fazliana Rahim et al.

Fig. 1. The Differences between Foreign Exchange Rate (FER) and the Forecasting
Method
Based on Fig. 1, the blue line denotes the Foreign Exchange Rate (FER).
The orange line denotes the forecast value using Weighted Subsethood-Based
Algorithm (WSBA) and the grey line denotes the prior forecasting method.
From the plotted graph, it was shown that the forecasting value of FER using
WSBA found to be nearly accurate to the Foreign Exchange Rate (FER). This
proves that the forecast value using FTS with the proposed of WSBA is better
compare to prior method and perform more precise forecasting of the FER.
Next, Table II below summarized the results of evaluation for each
forecasting method.
TABLE II: EVALUATION OF METHOD
Forecasting MSE RMSE Percent
Method Accuracy
Proposed WSBA 0.17 1.89 98.1 %
Prior Method 0.37 9.52 90.5 %

From the table III, the result shows the value of MSE and RMSE for
proposed Weighted SubsethoodBased Algorithm (WSBA) is lesser than the
prior method, which is 0.17 and 1.89 respectively. Meanwhile, the result of
MSE and RMSE for previous method is 0.37 and 9.52 respectively. Referring to
the percent accuracy also shows that forecasting using WSBA more accurate
compare to the prior method. Therefore, based on the above result, the
proposed WSBA can be used as a method in generating rule prediction of
time series forecasting.

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