This program uses a recursive neural network to predict future stock prices based on historical data. Users can input the ticker symbol of the stock, along with the quantity and buying price. The neural network will then use the Yahoo Finance API to fetch historical closing prices, which will be used to train the neural network. It will then predict the closing price a user-specified number of days into the future and give insight into the user's gain/loss in that period of time. It also displays a graph comparing the model's predicted prices to actual prices for days in the near past to show user's its accuracy. The model was developed with the help of a Youtube tutorial, and the ability to forecast any number of days into the future (with varying accuracy) was implemented by integrating ideas from many only sources. The program is delivered in a more user-friendly manner using a Flask web app which allows users to enter their input to see the graph and predictions.
amandachen1024 / stock-price-predictor-webpage Goto Github PK
View Code? Open in Web Editor NEWA neural network to predict future stock prices
License: Apache License 2.0