This is a challenge project which I completed to test my skills in Linear and Multiple Regression .
Goal of the Project:
Explore different parameters and build a perfect formula which to calculate the winnings
Overview of my approach
Started with a couple of linear regressions so find the best single parameter to predict the data Next, Tried and tested a few multiple linear regressions with 2 parameters And Finally came up with a model with 3 parameters to predict the data
I also concluded by comparing the efficiency of my model with the best suited parameters given my CodeCademy
Challenges faced:
I read the including parameter which is strongly correlated to some other existing parameter in your model would decrease the overall score because of redundency,So picking a pair of parameters which were strongly correlated to my outcome and not correlated within themselves was quite a challenge since a lot these were correlated with themselves.
Dataset used: tennis_stats.csv . This file is uploaded and was provided by CodeCademy
Feeback and suggestions are appreciated.