MAT_243_Project_Three_Summary_Report.docx MAT 243 Project Three Summary Report Souther

MAT_243_Project_Three_Summary_Report.docx    MAT 243 Project Three Summary Report  Southern New Hampshire University    1. Introduction  The purpose of this analysis is to assist in projections of up coming NBA game wins.  This will be achieved by accessing a large set of historical game data in which performance patterns will be analyzed.  The basis of the analysis will input performance metrics within regular games seasons in hopes to predict future wins.    2. Data Preparation  Within the database there are several variables that will be used to calculate the projections.These variables are identified as Average Points per game (ave_pts), Relative Skill of each team (avg_elo_n) and the Average Points Differential which is the difference between the team in question as well as the opposing team (avg_pts_differential).  The variables are relatively self-explanatory for a regular NBA season.     3. Scatterplot and Correlation for the Total Number of Wins and Average Points Scored  Correlation, by definition, is basically the association between two variables. Generally, if a variable increase as well as another variable increases a positive correlation is the result.  However, if one variable increases but the other decreases a negative correlation would be the result.  The coefficient correlation measures the strength between a predictorvariable and a response variable.    The scatterplot is a great visual to show either the positive or negative correlation.  Although it can look a bit overwhelming. It is quite easy