Abstract: Gold has always been considered to be the safest haven for investment by the Indians. In fact gold is widely regarded as a hedge against adverse financial and economic conditions by the Indians. Thus prior prediction of gold prices assumes great importance as it can aid both the investors and the traders in making intelligent investment decisions in the Gold market. Time Series forecasting has found wide spread applications in varied spheres of business, economics, commerce, production and many others. On the other hand, the past few years have also witnessed rising popularity of Artificial Neural networks for forecasting purposes. This paper compares and analyses the forecasting of the Indian Gold prices using the linear Autoregressive Integrated Moving Average (ARIMA) model and the non linear Artificial Neural Network (ANN) model by developing two models based on these approaches. The relative forecasting efficiencies of the two proposed models is then compared using the statistical measures of performance.

Keywords: Artificial Neural Networks, Autoregressive Integrated Moving Average (ARIMA), Indian Gold Prices, Time Series Forecasting.