5 November 2018
This code forecasts election-day results using a Bayesian structural time series model. Forecasts represent the most likely candidate to receive a plurality of votes.
The data is scraped from RealClearPolitics for informational and research purposes only.
- Florida: Gillum (D) (confidence: 92%)
- Georgia: Kemp (R) (83%)
- Kansas: Kobach (R) (62%)
Colored regions represent 95% credible intervals. Points represent observed polls, lines are estimated time-course for the true (unobserved) proportion of votes each candidate would receive, were the election held today.
