Resumen
Gross Regional Domestic Product (GDP) is the total number of products in the form of goods and services produced by production units within the boundaries of a country (domestic) for one year which is one important indicator to know the condition of an area within a certain period either at current or at constant prices. This research briefly wants to know the value or prediction (forecasting) from data of Gross Regional Domestic Product (PDRB) of Bangka Belitung Islands Province for the preventing action related to policy type which will be done by the decision maker. The method of forecasting used in this research is ARIMA method with the utilization of software R in the form of the value of significant t-test and by using parsimony principle consisting of two methods combined into one, namely AR (Autoregressive) and MA (Moving Average). This ARIMA model in its application is often written with ARIMA (p, d, q) whose descriptions p, d, q are the same as the previous ones. This method uses an iterative approach to the identification of an existing model. It can be seen that historical data have forecasting results that tend to have an uptrend as evidenced by the Q Ljung-Box test and ACF / PACF plot of data with the assay results that the residuals of the ARIMA model (1,1,0) are good models evidenced by plots ACF that there is no lag out of the interval line with the forecasting result that has the trend up from the third quarter of 2013 ie 11.028.917, 11.223.615, 11.360.718, 11.456.826, 11.523.985 and 11.570. 813. The results of this forecasting can certainly be applied for n years to come.