TAF PUBLISHING

Over the past few years, the modeling approach is widely used to discover the complex and dynamic relationship between various profitable variables. This paper focuses on the modeling of gold and silver rates concerning a certain period and predicting the gold and silver rates. In this paper, with the assistance of the Particle Swarm Optimization (PSO) algorithm, the BOX JENKINS and the Auto-Regressive Integrated Moving Average(ARIMA) models are developed for the considered economic variables. A comparative study is also presented to assure the model's accuracy. Following that, the PSO-based KALMAN FILTER DESIGN approach is implemented on the gold and silver rates to forecast market prices. In today's unpredictable world, investors believe that gold can act as a hedge against unexpected disasters, both natural and economic. Therefore forecasting the price of gold has been of the highest interest. The major advantage of the proposed PSO-based modeling and prediction approach is that it is a fully automated method which results in higher flexibility and greater accuracy. This study also confirms that the PSO-based ARIMA model yields a better result than the PSO-based BOX JENKINS model. The proposed PSO-based KALMAN FILTER approach also provides better predictions.