Using particle swarm optimization to a financial time series prediction

The financial system is generally a very complicated system. So, it is very hard to predict its data. For example, it is a hard work to prediction the stock market. In this paper a similar sequence matching (SSM) and a particle swarm optimization (PSO) proposed by author is combined to predict the financial time series. In order to verify this prediction methods, the stock market data of the Indonesia Stock Price Index in 1997 to 2010 is used. The result shown that to prediction the stock market trend is very good, but to predict the stock price it does not produce meaningful result.