Estimating player value based on Win Share and Salary data
This analysis utilizes linear regression to determine an NBA Player's Value based upon their on-court production.
I begin with raw data from basketball-reference.com and clean the data for analysis purposes.
From there, I begin to further understand the data through visual representation and statistical analysis.
Lastly, I utilize linear regression models to learn what features are most relevant and build a prediction model for players' salaries based upon their statistics from on-court performances.
The completed project files to review are: Capstone 1: NBA Player Value Based Upon On-Court Performance.docx Capstone 1: NBA Player Value Based Upon On-Court Performance.pdf (visuals) NBA_Player_Value_Capstone.ipynb