Tom wrote:
I have a database of many factors.
What would you suggest I do to start a regression study?
Look at single factors, combinations?
I could use Excel to plot finsh positions or beaten lengths of horses with certain factors by rank, look for postivie correlation?
##
Tom, check the above discussions on regression analysis. For a number of reasons this statistical approach is problematic in horse racing. In sum, unless you carefully control for multicolinearity (high correlation of variables) the results are suspect, certainly their Beta values would be. Also, you need to have probit or logistic curve as your dependent variable. Finally if you go on a fishing expedition throwing independent variables into the models you run the risk of interpretability problems. Remember that correlation is not causation. Correlation relationships will ultimately regress to the the mean and would make a poor basis for wagering.
Trying to save you some time and maybe ($'s)
|