/nflData

Some models and analysis to investigate efficiency of the NFL gambling market

Primary LanguagePython

nflData

#Script to implement a Logistic regression model of the form #Y^_i = b_0+ b_1HWP_i + b_2AWP_i + b_3HL4_i + b_4AL4_i b_5FAV_i + e_i #i is the index of the game number #Y^_i is our prediction for Y_i the indicator of the home team beating the spread in game i #{b_0,b_1,b_2,b_3,b_4,b_5} the parameters to train #HWP_i: overall win percentage (relative to the spread) of home team #AWP_i: overall win percentage (relative to the spread) of away team #HL4_i: number of times home team has beaten the spread in the last 4 games played (this season) #HL4_i: number of times away team has beaten the spread in the last 4 games played (this season) #FAV_i: indicator the home team is the favorite