Graphing Logits

Often one wants to get a quick evaluation of whether, at least approximately, a model assuming linearity on the logit scale is a reasonable assumption. One does this by graphing logits within groups.


We need first to understand how to calculate logits with groups.


The process can be accomplished using the SAS data step.


The process can also be accomplished using PROC SQL.


A Macro for Graphing Logits.



Homework
Use a copy of the dataset lipid2018 that includes the variables age and a numeric variable chd. The numeric variable chd is coded 1 if the original character variable equals "Yes" 0 otherwise.
Use the macro to graph the logits for age in 6 groups.


The slides used in the videos are found here