Open DavidPeleg6 opened 1 year ago
it should be able to produce an expected 'profit' for a given investment, the point is to have this thing show up in the streamlit UI to show users how much they could be making if they use our predictions
the proportion of investment in each stock can be equivalent to the rmse of the model predicting it. meaning we take the top 30 most buzzing stocks from sentiment, take the 5 top performing models on these stocks, and invest according to their rmse/r^2 rating.
Another direction: maybe the most buzzing stocks are the ones that are most likely to present new anomalous behavior that breaks the model. so in this case, lets test the same process but with the 30 least buzzing stocks
create a trading bot that will test what will be the profit if we use the model for trading (in a seperate .py file)