Meta-Stock-Price-Forecasting-Model

In this project, I had an opportunity to work with a complex dataset analyzing the historical stock data of Meta and creating a predictive model. This was my first FinTech project, and I had a great learning experience. In order to analyze the dataset, I started off with an extensive exploratory data analysis which helped understand the data before getting into fundamental and technical analysis. In fundamental analysis, I first calculated the earnings per share and then calculated the price to earnings ratio. After this I got into technical analysis by calculating moving averages and relative strength index. I plotted the graphs of these two features to better understand the trends. I then moved onto building a predictive model: I first used a Long Short Term Memory Neural Network model as it had a very high margin of error. In order to fix that, I used a Random Forest Regressor model to significantly reduce the root mean squared error and learnt that it is important to choose the right model to train data in such cases.