FORECASTING OF PRDB IN JEMBER, EAST JAVA USING LEVENBERG MARQUARDT METHOD

Forecasting of Product Domestic Regional Bruto (PDRB) in Jember has been developed to successful in consistently predicting the performance of the Domestic Government. The complex set of problem processes within PDRB have made the development of analytical models to be a significant challenge. Advanced simulation tools are needed to become more accurately model of PDRB systems. As an alternative approach, we have begun development of PDRB modeling based on Levenberg Marquardt which uses Matlab 7.6.0(R2008b). A neural network based learning system method has been proposed forecasting of PDRB. Levenberg Marquardt based technique is used for learning PDRB. Thus a precision model of Levenberg Marquardt has been evaluated. The correlation coefficient of this model is 0.99 shows good results for the target and network output.