Section 31 Multiple Linear Regression: Multiple Variables


31.1 Statistical Model

\[ \large y_{i} = \beta_0 + \beta_1 x_{1i} + \beta_2 x_{2i} + ... + \beta_p x_{pi} + \epsilon_{i} \]


31.2 Blood pressure data: Scatter plot


31.3 Important Questions

  • Are BMI, Age and other predictors useful to predict SBP?

  • What are the magnitude and direction of these effects?

  • How strong are the evidence?

  • Are there other predictors to explain the variability in SBP?

  • How many predictors will be useful to predict SBP?

  • How do we develop the model?

  • How to decide for predictors to be incorporated in the model?