No abstract

Artificial neural network (ANN) has advantage in time series forecasting as it has potential to solve complex forecasting problems. This is because ANN is data driven approach which able to be trained to map past values of a time series. In this study the forecast performance between neural network and classical time series forecasting method namely seasonal autoregressive integrated moving average models was being compared by utilizing gold price data. Moreover, the effect of different data preprocessing on the forecast performance of neural network being examined. The forecast accuracy was evaluated using mean absolute deviation, root mean square error and mean absolute percentage error. It was found that ANN produced the most accurate forecast when Box-Cox transformation was used as data preprocessing. High degree of accuracy can be yield on a wide range of forecasting applications since ANN is a flexible computing frameworks and universal approximates [6]. Although ANN able to produce reliable forecast, but mathematical proofs underlying the ANN will need to be considered in order to determine the best conditions to be

One of the issues that triggers worlds lately is the increasing rate of the unemployment rate. Consequently, this research objective is to compare the most accurate forecast method and to find the most suitable period to predict the future of Malaysia’s unemployment rate in 2016. There are five sets of Malaysia’s unemployment rate and three forecasting methods being used which are Naïve, Simple Exponential Smoothing (SES) and Holt’s method. The forecasting model was then selected based on the smallest accuracy measures. The results indicated that Holt’s is the optimal model in forecasting the overall yearly unemployment rate, male yearly unemployment rate and overall quarterly unemployment rate. Furthermore, for female yearly unemployment rate and overall monthly unemployment rate, the best forecasting method was SES. Meanwhile, the overall unemployment rate of Malaysia in year 2016 was predicted to be 2.9% while 3.4% was estimated to be the value of unemployment rate for second half year of 2016 by using quarterly and monthly data. The forecast value was remained the same as previous year for overall yearly male data and female data which were 2.9% and 3.3% respectively. Lastly, the best period in forecasting Malaysia’s overall unemployment rate was found to be month with the value of 3.4%.

No abstract

Forecasting crude palm oil price is important, particularly when the investors encounter with the increasing risks and uncertainties in the future. Therefore, the aim of this study is to forecast the price of palm oil in Malaysia for the next years based on price for the period of 31 years. The objective of the research is to propose an appropriate model to forecast the CPO price. Thus, this study proposes three types of models, which are namely: Autoregressive Integrated Moving Average (ARIMA), Autoregressive Conditional Heteroskedasticity (ARCH) and Generalized Autoregressive Conditional Heteroskedasticity (GARCH). Akaike Information Criterion (AIC) and Hannan-Quinn Criterion (H-Q) statistic were used to obtain the best model. It was found that ARIMA (2, 1, 5) performed better compared to ARCH and GARCH models. It is concluded that ARIMA (2, 1, 5) can be used as an alternative model to forecast the CPO price.