SAS - Logistic Regression

www.krohneducat...
This video describes the typical model used in logistic regression as well as how to perform an overall significance test, individual significance test, and determine if a reduced model is sufficient to replace a full model. This video also describes how to compute odds and probability for given predictor values.

[@saliht.3133]

I barely make comments on youtube. I have to say, this is one of the best videos I have watched on the concept. very straight to the point, and summarized.

[@TheHiddenPearl]

yes, his explanation is very clarity to understand, i am not sure why the sound is very quiet on my end I can barely hear.

[@maggie4003]

Echoing others. This video was, without a doubt, one of the best on You Tube - across any subject. Saved me a ton of work. Many thanks

[@shanemoe9098]

Thank you for your explanation Sir. SAS tutorials are hard to find and some of them are even confusing. But this video is very clear.

[@pullkitbbmp3715]

Thanks a lot, it is really helpful in understanding SAS logistic results. Thanks from bottom of my heart

[@pinkaura6983]

Question. Since Gender is a classification variable, unlike age which is continuous, wouldn't it be important to have gender in the class statement before the model statement?

[@onetwoBias]

Really helpful video, thanks a bunch dude!

[@maggie4003]

Interesting. I get similar, but different results using the same data set. Differences depend upon whether sex is classified in a Class statement, or considered numeric. I get identical odds ratio tests if sex is a "class" variable, although a different coefficient on sex (but same SE and p val). All other stats are the same IF I add a "class" statement for sex.

[@logophile92]

Can you please show the code that you did for the proc logistic procedure, though? How do you know what to put in the class statement? I mean, when you use a class statement in proc logistic, what does that mean?

[@qymyz_shubat]

Thank you very much for a good explanation, there is my question : isn't a partial test equivalent to likehood ratio, I mean the values of chi-square and p - value are the same

[@adriancobb6166]

Hiya, just a question, (I may of missed it). @ 16.16 with the odds equation where are you getting the e value from to get an answer of 23.39? Thanks!

[@JoelHernandez-vs1kn]

thanks

[@swatts0813]

Great Job!

[@pradeepmath]

well explained.

[@soumyopattnaik6787]

shouldn't Age be treated like a class or ordinal variable

[@briankwlin]

f-first :D

[@maggie4003]

Echoing others. This video was, without a doubt, one of the best on You Tube - across any subject. Saved me a ton of work. Many thanks