Stock Price Prediction using Machine Learning

It is a big challenge to predict how the stock market performs given various factors like people sentiments, fundamentals of the company etc. The stock prices of all the listed companies fluctuate everyday and it is very helpful if we can get an insight on how the stock has performed over the past years and predict the future price of the stock. In this project, the aim is to leverage 5 different machine learning models to compare the results of these models and choose the model with best results to predict future price of the stock. The results of the experiment show that LSTM model has the potential to predict the future price of the stock most accurately

Approach

1. Data Exploration

Dataset used: https://data.nasdaq.com/data/BSE/BOM500570-tata-motors-ltd-eod-prices
Stock: TATA MOTORS LTD.
Code: BSE/BOM500570

2. Data Inspection

3. Data Preprocessing (Handle Missing values, Sorting)

4. Data Visualization

5. Feature Selection

6. Split Data into Train and Test set (80:20)

7. Feature Scaling (MinMax Scaler)

8. Train Data on Machine Learning Models \

Linear Regression

KNN

SVM

Random Forest

LSTM

9. Evaluate performance using different metrics

10. Compare performance

11. Predict price for future dates