I need to design "A horse racing simulation" software. For that i need to analyze past records of races to determine a formula that can describe how the dependent variable Y respond to the changes in the independent variables/predictors x1,x2,x3...xN
My ideas are as follows: feed the predictors (stable efficiency, jockey efficiency, line of the horse, weight on the horse, distance of race, age of horse and rating of horse) and the dependent variable Y (finishing time) in a statistical package (SPSS|STATA) and obtain an accurate formula which relate them (Y=α+β1x1+β2x2+β3x3...+βNxN (N=number of variables)). Then i shall use the formula for upcoming races to get the values of Y in a race, finally sort the Ys (being the finishing times of all horses) and the least time = winner.
What techniques must i use to obtain the formula? Any suggestion or recommendation of books which might help me? Thank you :)
