bharatpurohit97 / stockprediction

Stock Prediction by Reinforcement Learning and RNN

StockPrediction

1. Yahoo Finance using Reinforcement Learning

Stock Prediction by Reinforcement Learning.

It's implementation of Q-learning applied to (short-term) stock trading. The model uses n-day windows of closing prices to determine if the best action to take at a given time is to buy, sell or sit.

As a result of the short-term state representation, the model is not very good at making decisions over long-term trends, but is quite good at predicting peaks and troughs

Usage

• To train the model :

cd ReinforcementLearning-YahooFinance
mkdir models
python train.py ^GSPC 10 1000`

• Then after training finishes :

python evaluate.py ^GSPC_2011 model_ep1000

Tutorial

Jupyter Notebook for stock prediction.

References

Deep Q-Learning with Keras and Gym - Q-learning overview and Agent skeleton code

Siraj Raval-School of AI

2. Google Stock Prediction Using Recurrent Neural Network

• Google Stock Prediction Using Recurrent Neural Network
• see plot in RNN-GoogleStock

stockprediction's Issues

no prediction, prediction is a data shift

in RNN-GoogleStock.ipynb you dont predict the next hour, you just take the price from last hour as your prediction. And in the end you shift all back by one and thats the reason why the accuracy is 100%

^GSPC_2011 Total Profit: $0.00

Thank you for this code.
I ran it for 160 epoch and then evaluated it. There is no result in the output:

python3 evaluate.py ^GSPC_2011 model_ep160
Using TensorFlow backend.
--------------------------------
^GSPC_2011 Total Profit: $0.00
--------------------------------

could you please publish some of the results.