A bunch of scripts that process stock market data and makes forecasts using popular statistical time series forecasting methods.
Proof that I passed university-level statistics (:
Every week:
- Stock data downloaded and cleaned - 3 months
- R calculates 1w forecast for each ticker and plots it
- R calcs previous model and gets accuracy, given latest observations.
- This accuracy data is then stored.
- Historical prediction accuracy is loaded from db and plotted.
- All graphs are published to the website.
- R
- ggplot2
- forecast
- fpp2
- python3
- mysql-connector-python