28nov2011 - CR Long C532: Regression Modeling 28 Nov 2011...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
CR Long C532: Regression Modeling 28 Nov 2011 1 Learning Objectives 1. To learn methods of selecting explanatory variable to include in multiple regression models. Multiple Linear Regression A flexible and widely used statistical tool for assessing the joint relationship of multiple explanatory variables with a continuous outcome variable. Purpose : Including multiple explanatory variables in a model allows to adjust for confounding variables (especially for observational designs), examining mediation, assessing interactions, and increasing precision of the estimates (especially for experimental designs). Terminology Confounding : 1 X causes Y , so must be controlled for in the analysis of 2 X and Y Mediation : 2 X lies on the pathway between 1 X and Y Review: Goodness-of-fit Statistics R : correlation coefficient between the observed and fitted values (0 to 1) R 2 : the proportion of the total variation that is explained by the linear regression Adjusted R 2 : corrects R 2 (decreases) for the fact that a model always fits the data it is
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 2
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 01/03/2012 for the course C 532 taught by Professor Long during the Fall '11 term at Palmer Chiropractic.

Page1 / 2

28nov2011 - CR Long C532: Regression Modeling 28 Nov 2011...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online