This project consists of four parts:
1.Google stock price prediction: Involves the use of LSTM to predict the price of google stocks and to analyze the findings. The data is obtained from yahoo finance and a comparison is drawn between actual price and predicted price of stocks. LSTM is a variety of recurrent neural networks (RNNs) that are capable of learning long-term dependencies, especially in sequence prediction problems.
2.Google stocks risk analysis: Involves the analysis of quantitative risk by the use of monte carlo simulation.By using this method , the distribution of a possible outcomes of events is generated by analyzing a modelseveral times, each time using random input values selected from the probability distribution considered normal of the components that comprise the model.
3.Generating buying and selling signals for given stocks: Involves the use of a combination of the MACD and RSI indicators to generate favourable long and short signals.MACD uses exponential moving averages which is a more enhanced version of simple moving averages(as weight is given to most recent data.RSI is based on average gains or losses during a lookback period(generally taken to be 14 days).
4.Stock recommender system: Involves the storing of stock price data in a sql database by making use of the sqlalchemy library available in python.This recommender system generates buying and selling signals after analysing the results of 3 technical indicators namely MACD,RSI and the golden cross.