I am trying to create a generalised linear model with random effect. I have a small dataset, with longitudinal data of 4 subjects. The data I obtain from them is a frequency data, and for one of the subjects all of the data points are 0. So when checking the normality and the residual plots the distribution is not normal.
I tried transforming the data in different ways but the plot remains to look the same.
Is there any model or transformation I can use for this type of data, where one of the subjects shows no variability?
m=lme(Freq ~ Time, random=~ 1|Subject, data=my_data, method='ML')
You have frequency data, with a small mean. You can't expect the residuals to look very normal. I am not sure why you are assuming a Gaussian model - you could try fitting using a Poisson family in a call to glmer() (package lme4).
If you really have zero inflated Poisson data, as TTNK says, you will need many more than 4 subjects.
The technical post webpages of this site follow the CC BY-SA 4.0 protocol. If you need to reprint, please indicate the site URL or the original address.Any question please contact:yoyou2525@163.com.