Hello,
I think there is a problem there.
You have Day and Month and Hour as independent random factors.
This means you assume that the first of January, the First of March and
the First of June have something in common.
The same thing is true for Hour, so you assume implecitely a daily
patter. You did not provide enough information to say what is the right
way of analysing it. Usually I woudl suggest to make a variable called
time from month, day and possibly hour values. Because it is a class
variable this makes only sense if you measure several points at the same
time (like the same hour).
You should also think that season can be confounded with the month effect.
Frank
On 21.5.2019 20.41, Manuel Spínola wrote:
Dear list members,
Sorry for cross-posting.
I measured an index on 12 points within each of 3 cover types on 4 seasons.
But the index was measured several times within hours, on different days
and different months.
I want to assess changes in the index among cover types and seasons.
Is the following model parameterization appropriate in the lme4 package:
mod_adi_01 <- lmer(ADI ~ Season*CoverType + (1 | Point) + (1 | Month) + (1
| Day) + (1 | Hour), data = df_01, REML = FALSE)
Manuel
_______________________________________________
R-sig-ecology mailing list
R-sig-ecology@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-ecology
