Link to report

The purpose of this project is to analyze baseball team data from 1871 to 2006 in order to predict the number of wins that a team will register in a regular season.

Multiple regression models have been used for the purpose. The models have been generated using variable selections techniques such as Forward, Backward or Stepwise selection.
The best model is selected and further analyzed to see if predictions can be made or not.
The data was a bit messy so it had to be cleaned first. Some exploratory analysis was also done.