3/17/2008

Discussion of Small Group Project

• Use the statistical methods and models we have discussed in class to analyze performance statistics for the athletes who have used or are accused of using performance enhancing drugs. Did these atheletes perform significantly better than their peers during the time period in question? Was the performance of these athletes significantly better during the time period in question than in other parts of their career?
  • Marion Jones
  • Floyd Landis
  • Roger Clemens
  • Mark McGwire
• See Mitchell_report.pdf
• Players named in the Mitchell Report

Simple Linear Regression - Example (p11-3.sas)

• Purpose
• Functional versus statistical relationship
• Formal Model
• Method of Least Squares
• Assumptions
  • Normality
  • Independence
  • Homogeneity of variance
• Estimation of variability about the regression line
  • Residuals
    • Residual = observed - predicted
  • MSE
• Assessing the fit
  • Residual plots (used to check assumptions)
    • Plot residuals against explanatory variable and other available variables
    • Construct a normal probability plot for the residuals
  • Coefficient of determination
• Inference for the population correlation coefficient

Please read Chapter 13 for class on Wednesday