Page 1 of 1

Validation of a model

Posted: Thu Dec 30, 2021 3:07 pm
by mhanu
Hi.
i am wondering if when i build a model let's say logistic regression in jamovi, and get all the metrics for that model like
Accuracy, Sensitivity, specificity AUC, the validation process (i.e. Cross-validation or Leave-one-out) has already been done or not?

in case not how can i perform a validation process on a model?

Thanks

Re: Validation of a model

Posted: Mon Mar 14, 2022 10:01 pm
by geofclobo
Similar to my recent post:

Is the deviance fit measure, tested by chisquare test, from logistic regression the same as shown at MiniTab documentation:
"Use the goodness-of-fit tests to determine whether the predicted probabilities deviate from the observed probabilities in a way that the binomial distribution does not predict. If the p-value for the goodness-of-fit test is lower than your chosen significance level, the predicted probabilities deviate from the observed probabilities in a way that the binomial distribution does not predict."
https://support.minitab.com/en-us/minit ... y-results/

If so, when i have a significant p<0,05 response on the oveall model test, i have to reject al my analisys?
Even if the Omnibus likelihood ratio tests of my factor variables shows significant efect?