In this competition, you will be modeling data from the 1992 US Presidential Election.  That election was one of the most memorable elections due in part to the success of a third party candidate – Ross Perot. The ‘Giant Sucking Sound’ refers to what Perot thought that the North American Free Trade Agreement (NAFTA) would do to jobs as they were sucked to Mexico.
To give you context, watch the short video
[‘The Perot Myth’](http://fivethirtyeight.com/features/the-ross-perot-myth/)
from Nate Silver’s
[FiveThirtyEight.com](http://fivethirtyeight.com/)
.
Data is provided on the locality (county/city) level. You will be modeling the Democratic votes (variable = ‘VOTE’) by locality.
The data has been split into two groups, a 'training’ set and a 'test’ set. For the training set, the outcome ( ‘VOTE’ ) is provided for each locality. You will use this set to build your model to generate predictions for the test set.
For each locality in the test set, you must predict the Democratic vote percentage (‘VOTE’). Your score is determined by the average of the absolute differences of your predictions and the actual values (Mean Absolute Error) in the test set.