Highlights
Politics:
Forecasting the 2020 US Presidential Election
Ongoing project to predict the outcome of the 2020 US Presidential Election. I created a poll-based forecast model that integrates demographic and public opinion data to predict how undecided voters will break on election day. The forecast uses a state-space model for poll aggregation and a Bayesian linear model to predict the undecided break.
Click the image or the “Politics” tab to learn more…
Science:
Tracking COVID-19 in the United States
Ongoing project to track daily new COVID-19 cases and deaths in the US, using unsupervised machine learning methods to better qualify geographical differences in the pandemic’s severity over time.
Click the image or the “Science” tab to learn more…
Music: