O'Reilly logo

Practical Predictive Analytics by Ralph Winters

Stay ahead with the world's most comprehensive technology and business learning platform.

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, tutorials, and more.

Start Free Trial

No credit card required

Fitting a GLM model

Now we will fit a logistic model using the glm package. Logistic regression typically predicts the probability of the 1 event, so make sure that the 1 events corresponds to what you want to predict. Often there is some confusion when variables are coded 1 for negative conditions such as not present, or does not exist. For our model, Pain=1 represents reduction of pain. Be sure to specify family="binomial" as an option, since that tells GLM that you will be running a logistic model:

PainGLM <- glm(Pain ~ Treatment + Gender + Age + Duration, data=df, family="binomial")  

The summary function will list the coefficients (Estimate), along with the standard error, standardized z value, and the p-values. The p-values (last column) ...

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, interactive tutorials, and more.

Start Free Trial

No credit card required