Cross Validated
2023-05-23 14:29 UTC
By Tom Johnson
AI-113-20230523-social-media-4142d5d4
Choosing a suitable GLM family for zero-inflated data non-integer data
I am currently aiming to find any statistical difference between site fidelity for four shark species between sex and sample site. This uses a mark-recapture method and thus the data is heavily zero inflated for sharks that were not recaptured. Since the values for site fidelity are not integers I am unable to fit a zero-inflated Poisson model using the "pscl" package. Are there any other options in terms of GLM model families that I could use, or are there other statistical tests that I could potentially use to find any differences between the groups?
I am currently aiming to find any statistical difference between site fidelity for four shark species between sex and sample site. This uses a mark-recapture method and thus the data is heavily zero inflated for sharks that were not recaptured. Since the values for site fidelity are not integers I am unable to fit a zero-inflated Poisson model using the "pscl" package. Are there any other options in terms of GLM model families that I could use, or are there other statistical tests that I could potentially use to find any differences between the groups?
Full article content could not be extracted automatically. Read the original below.
Source:
Cross Validated
· stats.stackexchange.com