2
$\begingroup$

I'm a master's level statistician and have been doing logistic regression for a while. I'm helping a friend who is taking an advanced stats course and ran across some terms I'm not familiar with when working with LR. One of the LR homework problems is asking for the model's proportional by chance accuracy rate, full model's accuracy rate, and the model's sensitivity rate. I'm not sure how the sensitivity rate is different than the accuracy rate. Also, the prof wants a PDA model on the same data set to compare to the LR. I'm not sure what a PDA model is or is this a term for something that is usually referred to as something else.

$\endgroup$
1
$\begingroup$

Sensitivity is the proportion of true cases which the model correctly identified as cases. Accuracy is the proportion of observations which were correctly classified (ie not just cases but non-cases). Note that the proportion of true non-cases correctly identified as non-cases is the specificity.

See also the Wikipedia article https://en.wikipedia.org/wiki/Sensitivity_and_specificity

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.