0
$\begingroup$

Ive made a binomial GLM in r and my resid. deviance and df are rather high;

Null deviance: 4396.2 on 3207 degrees of freedom Residual deviance: 3679.4 on 3205 degrees of freedom (4346 observations deleted due to missingness) AIC: 3685.4

Im aware that these values tell me my model isnt a good fit however is there a way i can change this? or do i just need to start from scratch and choose new X variables?

Also what is the AIC? (if you cant tell im new to GLMs and stats in general) lol

$\endgroup$
  • $\begingroup$ You don't interpret the deviance values by themselves, you compare the null and residual deviances. If the null and residual deviances are similar then your model is not really much better then the null one. In your case, it isn't. Show us your code and what you have done so far. You may need to add more variables in your model. $\endgroup$ – user2974951 Dec 20 '18 at 14:29
  • $\begingroup$ mod2<-glm(prevbygiemsa~Schisto.elisa+Sex, family = binomial(link = logit),na.action=na.exclude, data = dframe1), thats the code i used. should i try adding more variables? $\endgroup$ – Dean Burchell Dec 20 '18 at 16:14
  • $\begingroup$ So how many observations do you have in your data set? I see that you have over 4000 missing values. Do you have any other variables that you could use? $\endgroup$ – user2974951 Dec 20 '18 at 18:08

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.