Abstract
Application of Feature Selection and Extraction techniques on Indian Stock Market
by
Saif Jawaid
The analysis of the financial market always draws a lot of attention from investors and researchers. The trend of stock market is very complex and is influenced by various factors. Therefore, to find out the most significant factors to the stock market is very important. Feature Selection is such an algorithm that can remove the redundant and irrelevant factors and figure out the most significant subset of factors to build the analysis model. In this paper, 8 stocks of 8 different industries were analysed. Based on the results given by the 5 different feature selection algorithms a score system was developed and features with scores of more than 2.5 were chosen for further analysis. A comparison of the absolute error was made between features chosen based on company specific score vs features chosen based on composite score. Further, new features were extracted from the features selected based on the composite. An absolute error comparison was made between features selected based on composite score, both, with and without the newly extracted features.
Committee
Workshop
Key Dates
Communication
StatFin Main Webpage