Case Study: Cornell University Johnson Graduate School of Management achieves refined 2014 U.S. Senate control probabilities with Palisade's @RISK

Cornell Professor Takes Nate Silver Analysis of U.S. Senate Midterms a Step Further using Monte Carlo Simulation

Cornell University Johnson Graduate School of Management Professor Lawrence W. Robinson used Palisade’s @RISK to extend Nate Silver’s FiveThirtyEight 2014 U.S. Senate forecasts by adding Monte Carlo simulation. His challenge was to compute the probability that Democrats would hold at least 50 Senate seats while accounting for varying degrees of correlation between races—a task that is cumbersome or essentially impossible to do by hand for intermediate correlation levels.

Using Palisade’s @RISK, Robinson ran thousands of correlated and uncorrelated simulations to produce lower and upper bounds (41%–50%) and a full probability curve across correlation coefficients; he showed that if correlations fall between 20% and 85% the Democrats’ chance is about 45% ± 0.19. Palisade’s @RISK made the complex intermediate-correlation calculations feasible, produced clear, measurable probabilities, and provided a timely classroom and analytical demonstration of Monte Carlo’s value.

Cornell University Johnson Graduate School of Management

Lawrence W. Robinson

Professor of Operations Management