3
$\begingroup$

I'm a complete novice when it comes to stats, so my question may not even be relevant. I have been using the glm() function in R to model a response variable (bird species) versus predictor variables (habitat variables etc.) with a Poisson distribution. I selected the best model based on the AIC. Now I'm curious if I can visualize the results in some way. Basically, can I plot the response versus the predictor variables? I want to try and figure out the optimum habitat variables for each bird species. For example: let's say Yellow Warbler had a positive relationship with Shrub Height... but what is the optimal height??

$\endgroup$

put on hold as unclear what you're asking by mkt, Michael Chernick, Sycorax, whuber yesterday

Please clarify your specific problem or add additional details to highlight exactly what you need. As it's currently written, it’s hard to tell exactly what you're asking. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

  • $\begingroup$ Have you stumbled upon this tutorial? $\endgroup$ – Antoni Parellada Feb 1 '17 at 21:56
  • 1
    $\begingroup$ The big question is how you're modeling this. If you are modeling the relationships as linear (in the space of $\log(\hat\lambda_i)$), the line has to go up (or down) as you move along any predictor variable. Therefore, the 'optimum' has to be infinity, negative infinity (or 0 or some other bound as appropriate). $\endgroup$ – gung Feb 1 '17 at 22:03
  • 1
    $\begingroup$ It'd surely help if you could give more details about your model. But for the sake of brevity, take a look at the sjPlot package, function sjp.glm $\endgroup$ – Diogo B Provete Feb 2 '17 at 12:36
  • $\begingroup$ Thank you, Antoni, I had seen that tutorial and I will try to work though it more. Here is an example of the code I am using:AMGOmod.1 <- glm(AMGO ~ mowtime + VegHgt.Avg + VegHgt.Stdv + AvgCanopy + Avg.TreeHgt + Avg.ShrubHgt + Avg.ShrubWdt + ShrubSTM + TreeSTM, data=datastepreg2016, family = poisson) $\endgroup$ – Lisa Feb 2 '17 at 16:28
  • $\begingroup$ Thanks so much Diogo!! I'm not positive (again, excuse my lack of stats knowledge) but the sjp.glm function seems to be what I was looking for! $\endgroup$ – Lisa Feb 2 '17 at 17:33