Sri Lankan Gold Price Forecasting - Using Artificial Neural Networks (ANN) Abstract
According to Dr Kennedy D. Gunawardene in 2009 The Artificial Neural Network is a collection of simple processors connected together and Each processor can only perform a very straight forward mathematical task, but large network of them has much greater capabilities and can do many things which one of its own can’t. The aim of this study is to find a model for forecasting gold prices in Sri Lanka by using artificial neural network. Data were collected from the market information during the period of January to February of 2006 and this experiment shows a sufficient prediction method in gold prices forecasting. Introduction
In recent years, the gold prices of Sri Lanka have been changed in considerable amounts. It shows some times gold prices increase and sometimes gold prices decrease. Because of that, the future fluctuations in gold prices in Sri Lanka are helpful to investors, businesses and general community. Gold is not being madding by Sri Lanka. This product is being imported from the other countries. So, many factors affect to gold prices in Sri Lanka. In this study, the gold price of one ounce of 24 karat is considered. Based Research Papers
In this study I based the research “A Neural Network Model for Gold Market” by Peter J. MaCann and Barry L. Kalman in department of computer science of Washington University. In this research artificial neural network has been used and ten inputs have been used. They are standards and poor 500 index, dollar index, bond index, bond yields, sterling currency index and gold mining index. And in this research five hidden layers have been used. There were two output variables as sell or buy. This based research has found there is a significant relationship between above inputs and prices of gold. The buying decision and selling decisions regarding the gold can be made by using this...