Abstract / Synopsis
We develop a simulation model for predicting the outcome of the US Presidential election based on simulating the distribution of the Electoral College. The simulation model has two parts: (a) estimating the probabilities for a given candidate to win each state and DC, based on state polls, and (b) estimating the probability that a given candidate will win at least 270 electoral votes, and thus win the White House. All simulations are coded using the high-level, open-source programming language R. One of the goals of this paper is to promote computational thinking in any STEM field by illustrating how probabilistic modeling and computer simulations can solve real-world problems for which analytical solutions may be difficult to find.
Boyan Kostadinov, "Predicting the Next US President by Simulating the Electoral College," Journal of Humanistic Mathematics, Volume 8 Issue 1 (January 2018), pages 64-93. DOI: 10.5642/jhummath.201801.05. Available at: https://scholarship.claremont.edu/jhm/vol8/iss1/5