1
$\begingroup$

Is the use of the term "Predictors" with regard to IVs in a logistic regression model appropriate or should an alternative term, such as correlates, be used instead?

Specifically thinking of a cross-sectional design.

$\endgroup$

1 Answer 1

0
$\begingroup$

Depends on what one is doing with it. For example. for multinomial logistic regression one might desire to solve for the individual probabilities, with other possabilities being a set of independent binary regressions, estimating the coefficients, or as a latent-variable model.

One could solve for correlations, but I would take that from the logarithm of a power formula ($Y=X_1^a\;X_2^b\;X_3^c\dots$) to do that (Well, sometimes I would, when it's appropriate).

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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