topic test statistics on odds ratios in logistic regression in Discussions https://community.jmp.com/t5/Discussions/test-statistics-on-odds-ratios-in-logistic-regression/m-p/10297#M9992 <HTML><HEAD></HEAD><BODY><P>I've been using logistic regression with multiple categorical predictors.&nbsp; I'm particularly interested in the odds ratios, and I understand how the odds ratios themselves are calculated, but I'm having trouble understanding how the test statistics associated with each odds ratio are calculated.&nbsp; JMP gives the disclaimer "<SPAN style="font-size: 10pt; line-height: 1.5em;">Tests and confidence intervals on odds ratios are likelihood ratio </SPAN><SPAN style="font-size: 10pt; line-height: 1.5em;">based."&nbsp; My question is, where does the p value associated with each odds ratio come from?&nbsp; Are they derived from a likelihood ratio?&nbsp; As I understand it, the whole model test, goodness of fit test, and effects tests all use the likelihood ratio to compute a chi square test statistic, but I cannot figure out how the p value (and the confidence intervals) for a given odds ratio are calculated.&nbsp; I've searched extensively online for an answer to this question with no success.&nbsp; It seems that if the explanatory variables are continuous, there is no p value associated with the odds ratio.&nbsp; In my case, the explanatory variable is categorical, with 3 levels, and there are odds ratios for each pairwise comparison.&nbsp; Any help with this would be much appreciated!<span class="lia-inline-image-display-wrapper lia-image-align-inline" image-alt="7736_Fit Model.jpg" style="width: 456px;"><img src="https://community.jmp.com/t5/image/serverpage/image-id/1050iB172EBC548A432EE/image-size/medium?v=v2&amp;px=400" role="button" title="7736_Fit Model.jpg" alt="7736_Fit Model.jpg" /></span></SPAN></P></BODY></HTML> Tue, 18 Oct 2016 22:10:57 GMT mjahnke 2016-10-18T22:10:57Z test statistics on odds ratios in logistic regression https://community.jmp.com/t5/Discussions/test-statistics-on-odds-ratios-in-logistic-regression/m-p/10297#M9992 <HTML><HEAD></HEAD><BODY><P>I've been using logistic regression with multiple categorical predictors.&nbsp; I'm particularly interested in the odds ratios, and I understand how the odds ratios themselves are calculated, but I'm having trouble understanding how the test statistics associated with each odds ratio are calculated.&nbsp; JMP gives the disclaimer "<SPAN style="font-size: 10pt; line-height: 1.5em;">Tests and confidence intervals on odds ratios are likelihood ratio </SPAN><SPAN style="font-size: 10pt; line-height: 1.5em;">based."&nbsp; My question is, where does the p value associated with each odds ratio come from?&nbsp; Are they derived from a likelihood ratio?&nbsp; As I understand it, the whole model test, goodness of fit test, and effects tests all use the likelihood ratio to compute a chi square test statistic, but I cannot figure out how the p value (and the confidence intervals) for a given odds ratio are calculated.&nbsp; I've searched extensively online for an answer to this question with no success.&nbsp; It seems that if the explanatory variables are continuous, there is no p value associated with the odds ratio.&nbsp; In my case, the explanatory variable is categorical, with 3 levels, and there are odds ratios for each pairwise comparison.&nbsp; Any help with this would be much appreciated!<span class="lia-inline-image-display-wrapper lia-image-align-inline" image-alt="7736_Fit Model.jpg" style="width: 456px;"><img src="https://community.jmp.com/t5/image/serverpage/image-id/1050iB172EBC548A432EE/image-size/medium?v=v2&amp;px=400" role="button" title="7736_Fit Model.jpg" alt="7736_Fit Model.jpg" /></span></SPAN></P></BODY></HTML> Tue, 18 Oct 2016 22:10:57 GMT https://community.jmp.com/t5/Discussions/test-statistics-on-odds-ratios-in-logistic-regression/m-p/10297#M9992 mjahnke 2016-10-18T22:10:57Z