Stock price prediction using RNN

A Recurrent Neural Network using time series data for stock price prediction.

Data

To train the stock price prediction model the stock closing prices for the S&P500 were considered.

Model

A Recurrent Neural Network (RNN) is a type of neural network well-suited to time series data. To develop and train the model Tensorflow have been used in Python.

Evaluation

A test set has been used for evaluation. Here, each three days, the stock closing price for the next three days was predicted by the model.

• MAE: 55.79
• RMSE: 67.67