A Comparative Study of NIFTY-50 Prediction Using Linear Regression and Regression Tree Models

It is important for investors to predict the tendency of stock. However, the stock market fluctuation is too hard to predict. A single stock may be inflected by many elements, so it would be much more feasible to study a stock index. A stock index may be decided by many stocks, which means that it would not be effected by some only one company. In this passage, NIFTY-50 would be analyzed to make a prediction. The linear regression and regression tree are used to predict the close price of NIFTY-50. To make sure the data size is big enough, the data includes the price from 2000 to 2021. By comparing two models, a proper model to predict stock index would be found. After the comparison of the MSE and R square index, it is easy to find that the regression tree is the model much more reliable. However, there is still long way to go before researchers find way to predict stock price precisely.