Analysing the Fluctuations in Commodity Prices and Forecasting the Future Directions Using Machine Learning Techniques

India produces, imports, and exports many commodities. Of course, the different sectors of these commodities, as well as their stocks, are interdependent. The stock market also depends on other aspects such as the country’s economy, geographic and political conditions. In 2020, the coronavirus pandemic hit, leading to lockdowns across most of the world. Then in early 2022, the war between Russia and Ukraine has caused a significant disruption in the supply and demand chain and these are just two of the many factors affecting Indian market stock prices. This research focuses solely on analysing the fluctuations in the stock prices of pharma sector and oil sector. From each sector one stock is picked for analysis. These are Dr. Reddy’s Laboratories Limited from sector and Gujarat Gas Limited form commodity sector. One is a pharmaceutical company that manufactures and markets generic formulations, while the latter is a leading natural gas production company. Furthermore, this work will also predict the future directions of these stocks (next day price) by using three machine learning models such as LSTM, Random Forest Regression and LASSO Regression. The experimental results show that the Random Forest model performs better than the other two models with a lowest MAPE value of 0.75 for Dr. Reddy and 1.47 for Gujarat gas respectively.