PyStock is a python package that scans stocks and predicts prices.

Browse to the directory where this file lives, and run:

pip install .

This command will download any dependencies we have.

• Find abnormally high volume stock
• Get technical data: OHLC, pct_change, volatility
• Quantify fundamental data: EPS, Market cap, Revenue growth, etc. (used to filter high volume stocks)
  • Is company growing?
  • Is it in an innovative sector?
• Media attention over some reference period: positive/ negative catalysts?
• Create risk metric for the stock (0 low to 1 high- maybe with sigmoid function?)
  • The more days it goes green, the higher the risk.
  • The number of days elapsed is from the day the volume exploded.
  • Should be weighted somehow by news.
• Predict probability that price will go up on next market open.

• add ma50, ma200
• Filter high vol stocks based on fundamental data
• Attention model for timeseries forecasting