A Novel Linked Clustering Technique for Stock Market Predictions

S. Punitha and M. Jeyakarthic

Most people invest in stocks for high returns within a specific period of time. Stock values keep changing every day and even every hour due to various factors like political, economic, social and even psychological factors. Complicated inter-relationships between these factors result in price fluctuations which makes it challenging to judge or conclude end results manually. Although various techniques for their predictions are available, there is scope for more and better analysis of volatile stocks. Hence an efficient technique is proposed for predicting stock values called LCTSMP which can be adapted to ending stock prices. The results of the proposed technique, applied on Istanbul Stock Exchange Data, show that it can be applied to any volatile data like stock prices within a specified timeframe.

Volume 11 | 11-Special Issue

Pages: 833-839

DOI: 10.5373/JARDCS/V11SP11/20193104