Forecasting 2020 U.S. Presidential election results using the smoothing mixed effects model and spline extrapolation method proposed in Wright 2018. An extension of my final MS Statistics project that used data from primary elections (see primary-sme-masters under wilsonlisa/course-projects).
Shiny app: forecasts and poll tracker
- Correctly predicted the winner in 47 of 51 states + DC
- Miscalled AZ, FL, and GA (but predicted a difference of less than 1 percentage point in FL)
- Predicted the difference between Biden and Trump percentages of votes within 1 point in PA, WI, and WY
generalsme.R contains bulk of the code.
Wright, F.A., Wright, A.A., 2018. "How surprising was Trump's victory? Evaluations of the 2016 U.S. presidential election and a new poll aggregation model." Electoral Studies. 54, 81-89.