You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I have a dataset that i am struggling with how to approach handling. I have >1500 animals, with 2 samples per individual. The samples come from 2 places on the claw, thus trying to get a seasonal component. However, Im not seeing if there is a way to deal with paired samples unless individual is included as a random effect. I see the alligator example, but I don't have a "continuous" variable to account for a time change between samples.
I was hoping to assess the differences between the two segments of the claw and then if they are not capturing significantly different diets, I only need to use 1 which would reduce the sample size from ~3,000 down to ~1500 and then I could subset out geographic, sex, age, etc. With random effect being ~1500 individuals, this is going to take an exorbitant amount of time to run. Wont it? it will have to create a model for each individual? Am I missing other ways to deal with paired samples? Is there a way to modify the jags model to account for that?
Suggestions on how to deal with paired samples would be extremely helpful
The text was updated successfully, but these errors were encountered:
I have a dataset that i am struggling with how to approach handling. I have >1500 animals, with 2 samples per individual. The samples come from 2 places on the claw, thus trying to get a seasonal component. However, Im not seeing if there is a way to deal with paired samples unless individual is included as a random effect. I see the alligator example, but I don't have a "continuous" variable to account for a time change between samples.
I was hoping to assess the differences between the two segments of the claw and then if they are not capturing significantly different diets, I only need to use 1 which would reduce the sample size from ~3,000 down to ~1500 and then I could subset out geographic, sex, age, etc. With random effect being ~1500 individuals, this is going to take an exorbitant amount of time to run. Wont it? it will have to create a model for each individual? Am I missing other ways to deal with paired samples? Is there a way to modify the jags model to account for that?
Suggestions on how to deal with paired samples would be extremely helpful
The text was updated successfully, but these errors were encountered: