How to output the standard error of prediction in PROC logistic?

[deleted_user]

In the output of PROC logistic, I can get the upper (UCL) and lower (LCL) confidence limits of the predicted probability. Can I use the formula (UCL-LCL)/(2*1.96) to get the standard error of the prediction? It seems to me that the standard error produced by this formula is way smaller than I would expect.

[deleted_user]

when you do logistic regression, you get ODD RATIO. check http://slack.ser.man.ac.uk/theory/association_odds.html. You have to log your CI first.

[StatDave]

The confidence limits for the predicted probabilities are computed as shown in the "Details: Linear Predictor, Predicted Probability, and Confidence Limits" section of the LOGISTIC documentation:

http://support.sas.com/documentation/cdl/en/statug/63033/HTML/default/logistic_toc.htm