Machine Learning Models in Stock Market Prediction

The paper focuses on predicting the Nifty 50 Index by using 8 Supervised Machine Learning Models. techniques used for empirical study are Adaptive Boost (AdaBoost), k-Nearest Neighbors (kNN), Linear Regression (LR), Artificial Neural Network (ANN), Random Forest (RF), Stochastic Gradient Descent (SGD), Support Vector (SVM) and Decision Trees (DT). Experiments based historical data of Indian Stock Market from 22nd April, 1996 to 16th 2021, which is time series around 25 years. During period there were 6220 trading days excluding all non days. entire dataset was divided into 4 subsets different size-25% data, 50% 75% data. Each subset further 2 parts-training testing After applying 3 tests- Test Training Data, Testing Data Cross Validation each subset, prediction performance models compared after comparison, very interesting results found. evaluation indicate that Boost, k- Nearest Neighbors, under performed with increase in size set. shown almost similar among but took more training validating model. Thereafter better rest set, than Machine.