Season Statistics with Points Kaitlyn Kramer, Lauren Johnson - - PowerPoint PPT Presentation

season statistics with
SMART_READER_LITE
LIVE PREVIEW

Season Statistics with Points Kaitlyn Kramer, Lauren Johnson - - PowerPoint PPT Presentation

Associating NBA Season Statistics with Points Kaitlyn Kramer, Lauren Johnson Villanova University Variables Points per game Opponent points per game Offensive rebounds per Opponent offensive game rebounds per game


slide-1
SLIDE 1

Associating NBA Season Statistics with Points

Kaitlyn Kramer, Lauren Johnson Villanova University

slide-2
SLIDE 2

Variables

 Points per game  Offensive rebounds per

game

 Defensive rebounds per

game

 Assists per game  Steals per game  Blocks per game  Turnovers per game  Conference: East vs. West  Opponent points per game  Opponent offensive

rebounds per game

 Opponent defensive

rebounds per game

 Opponent assists per game  Opponent steals per game  Opponent blocks per game  Opponent turnovers per

game

slide-3
SLIDE 3

Objectives

 Observe relationships between season averages of

points per game based on other regressor variables

 Assess the validity of a final model using selected

regressor variables

 Fitting model to test data before estimating the final

coefficients

slide-4
SLIDE 4

Cross-Validation

 Had 1086 observations in full data set  Removed 10% of the observations (109) to set aside

for test data

 977 observations used in training data  Test data will be used to asses validity of the final

model

slide-5
SLIDE 5

Analysis of Outliers

slide-6
SLIDE 6

Response vs. Regressors

Opponent offensive rebounds per game, opponent defensive rebounds per game and opponent assists per game also displayed non-linear relationships with points per game.

slide-7
SLIDE 7

Regressors vs. Regressors

slide-8
SLIDE 8

Re-expressions

slide-9
SLIDE 9

Variable Selection

 Stepwise  Collinearity  Entry significance level of 0.10  Exit significance level of 0.05

Final Model:

PTS~OppPTS+AST+OppAST+OppORB+TOV+STL+BLK

slide-10
SLIDE 10

Test Data

r=0.91 R2=0.81

slide-11
SLIDE 11

Final Model

slide-12
SLIDE 12

Conclusions

 The data behaved well in terms of what objectives we were

trying to reach

 Residual analysis did not give us any trouble when testing

for homoscedasticity

 Original assumption that there would be more team stats in

final model was wrong

 Initial concerns of problems with collinearity between all of

these variables were handled through the use of stepwise variable selection

 Were able to reach our goal of understanding more about

the relationships between points per game and other game stats