I have a dataset with distances between beneficiaries and the nearest provision point (nearest hub).
I want to develop a model to explain distances based on several atrributes like category of beneficiary, category of provision point among others:
distance ~ cat_beneficiary + cat_provision + altitude + ...
I guess I should use a GLM, but I don't know which model would fits better with this kind of data (continuos and positive). Can I use a count data model (like Poisson or NB)? Or they just work with discrete data?
I attach a histogram.
I was researching about it and I think the best option is trying a Gamma family distribution. The only problem is that my data has some zeros. However, I replaced by 10^-6 which is not a classy solution but it worked fine.
Interesting question. Is it related to logistic optimisation for supply network ?
No, the research is about provision points and beneficiaries of ecosystem services.
Dear Josep,
Replacing zeros with small numbers (10^-6 or .Machine$double.xmin) may "solve" the problem, but a more elegant solution is using zero-inflated models (e.g. packages 'glmmTMB' or 'glmmADMB').
HTH,
Ákos
Thanks, Ákos Bede-Fazekas , for your advise. However, I tried the function and it still gives an error due to zeros:
> glm.dist
Hello Josep,
Hm.. nice... Maybe "zi=~1, family= ziGamma" solves the problem, but it is a bit strange that a zero-inflated model can not accept zeros...
Sorry for the wrong suggestion,
Ákos
Thank you for your efforts. The model (replacing zeros with Machine$double.xmin ) presenta un R2 de 0.12. Therefore, we decided to discard it because it is not able to explain the variance but in a small percentage.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Regression