Using methods from stochastic calculus, brownian motion, and statistical mechanics, we have developed algorithms to predict the outcome of the 2014 US Senate Midterm elections. You can find our results below. Note, the raw data was obtained from http://elections.huffingtonpost.com/pollster, a great service provided by Huffington Post. We used MATLAB to implement the algorithms and create the output. These algorithms were completed under ISK Analytics Inc. and the data extraction and research was done by Hargun Singh Kohli
These graphs show the probability of each candidate winning the November, 2014 elections.
The graphs above were generated by our stochastic calculus algorithms, in which we generate 100,000 random walks to obtain the predictions above. We show below the output of these random walks on election day. For brevity and clarity, we have shown the results of the first 100 random walks, but the pattern is clear.
The results of our random walk experiments based on our stochastic calculus algorithms. These results show the candidate’s popular vote percentage on election day, November 4, 2014.
The output summary generated by our algorithm is as follows:
Relativity Digest 
One reply on “2014 US Senate Midterm Elections Predictions”
Nice simulations here, but I’m curious if you would be willing to share more detail about the methods, and in particular the algorithms you used?