Is Joe Biden's Twitter More Likely to Move Markets Than Trump's?

We took this database of tweets we try to identify the words that occurred more frequently in those that move to market. And then we built a random forest model that tried to account not only for the relatives or value of each of those words and categories in generating market moves but also the interactions between them. So, if us if good appeared with the word China that had a different meaning than ifgood appeared with with a different word. It had statistical significance in modeling volatility which is important because it helps explain drivers of interest rate volatility. We incorporate this index into that that statistical explanation. It plays a significant role.