1 / 6

Multiple Regression

Multiple Regression. Similar to simple regression, but with more than one independent variable R 2 has same interpretation Residual analysis is similar Confidence & Prediction Interval are similar. Multiple Regression.

Download Presentation

Multiple Regression

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Multiple Regression • Similar to simple regression, but with more than one independent variable • R2 has same interpretation • Residual analysis is similar • Confidence & Prediction Interval are similar

  2. Multiple Regression • A multiple regression model includes a coefficient for each independent variable • Simple case is a quadratic model on a single variable • Independent variable can be indicator (dummy) variable • i.e. gender = 0 for female and gender =1 for male • Coefficients are called “partial slopes”

  3. Multiple Regression • A multiple regression model includes a coefficient for each independent variable • Collinearity occurs when two or more independent variables are correlated, thus explain the same information • Model can include interaction terms if independent variables are interact

  4. Variable Selection • Several procedures have been developed for selecting the best model for predicting Y from several independent variables (X’s) • Compare all possible regressions • Backward elimination • Forward Selection • Stepwise Elimination

  5. Logistic Regression • A regression model with a qualitative (typically dichotomous) dependent variable • Dependent variable can be thought of as a binomial response • i.e. Y=1 if patient is cured, and Y=0 otherwise • Model is constructed to predict P(Y=1) using a logistic function

  6. Logistic Regression • Linear relationship between the natural log of the odds ratio and the independent variables. • Odds ratio is the ratio of probabilities of success to failure • Each coefficient describes the size of the contribution of that “risk factor”

More Related