0
$\begingroup$

I have the following results from a logistic regression with one categorical predictor with two levels, fitted using the phyloglm() function from the phylolm package to account for the phylogenetic relationships in the data:

Call:
phyloglm(formula = count ~ predictor, data = df, phy = tree, 
    method = "poisson_GEE")

Method: poisson_GEE
Parameter estimate(s):
scale: 0.2965808 

Coefficients:
            Estimate   StdErr z.value   p.value    
(Intercept) 0.992002 0.077906 12.7334 < 2.2e-16 ***
predictorB  0.200457 0.058405  3.4322 0.0005987 ***
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Note: Wald-type p-values for coefficients, conditional on scale=0.2965808

Am I correct in interpreting the coefficient estimate for predictorB as the difference between predictorA and predictorB corresponding to a exp(0.200457) = ~1.22 = a ~22% increase in the expected value of count?

$\endgroup$
  • 1
    $\begingroup$ Yes. This is correct. You can always use the equation to make predictions for both groups then clarify the percentage increase. $\endgroup$ – Heteroskedastic Jim Jul 29 '18 at 9:13

Your Answer

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

Browse other questions tagged or ask your own question.