JOINPOINT REGRESSION ANALYSIS AND AN APPLICATION ON ISTANBUL STOCK-EXCHANGE

Joinpoint Regression Analysis is one of the statistical methods used to identify the best-fitting points if there is a statistically significant change in the trend. The aim of this study is to apply joinpoint regression analysis in the stock market and compare the performance of this method according to actual data set and estimated values. For this purpose, we collected the data set from the National Istanbul Stock Exchange (ISE) 30 index for April-May 2013 and examined that data set via Joinpoint Regression Analysis. We applied linear and nonlinear techniques with the help of Joinpoint software and determined the best technique according to their Mean Square Errors (MSE). With the projection for the future months and the actual results, we see that the estimated values are a little higher than the actual values However, this shows that we may apply Joinpoint regression to a time series data set in order to forecast future values.