2
$\begingroup$

I am analysing a data of number of aphids with variation over time (32 avaliations performed every week), area (two areas), and plant position (fixed factor), with 3 replications. Because of that, I am running both GEE and GLMM (to compare the results). The problem is that for each time in each area I have the data on precipitation and temperature. And these data are important because the aphid population is highly affected by these ambient conditions. I really want to study the effect of temperature and precipitation, and I am thinking of using them as fixed factors, but it does not sound correct because they are a range of values. I have thought about using the values as an offset in the model, but I don't know if it is right and probably I will not be able to make any conclusions on how the temperature and precipitation affected the data.

Here is a small example of the data set:

  • REP- repetition
  • posi - plant position
  • prec - precipitation
  • medt maxt mint - temperature information
  • area- area
  • cont- number of aphids
  • time- week of the count
rep   posi prec   medt    maxt    mint   area cont time



1 s   11.03   21.60   28.20   15.10   1   22     1

1 m   11.03   21.60   28.20   15.10   1   0      1 

1 i   11.03   21.60   28.20   15.10   1   0      1

2 s   11.03   21.60   28.20   15.10   1   2      1

2 m   11.03   21.60   28.20   15.10   1   0      1

2 i   11.03   21.60   28.20   15.10   1   0      1

3 s   11.03   21.60   28.20   15.10   1   1      1
.   .     .       .       .        .     .   . 

.   .     .       .       .        .     .   . 

1 s   143.70  22.66   26.24   19.09   2   0      32

1 m   143.70  22.66   26.24   19.09   2   23     32

1 i   143.70  22.66   26.24   19.09   2   18     32

2 s   143.70  22.66   26.24   19.09   2   0      32

2 m   143.70  22.66   26.24   19.09   2   0      32

2 i   143.70  22.66   26.24   19.09   2   0      32

3 s   143.70  22.66   26.24   19.09   2   3      32

3 m   143.70  22.66   26.24   19.09   2   0      32

3 i   143.70  22.66   26.24   19.09   2   3      32
Improve this question
$\endgroup$
3
  • $\begingroup$ This is hard to follow. Can you provide a small example dataset to illustrate your situation? $\endgroup$
    – gung - Reinstate Monica
    Commented Jul 28, 2017 at 14:18
  • $\begingroup$ I have edited this for English grammar based on my interpretation of what you are saying. Please ensure it still says what you want it to. $\endgroup$
    – gung - Reinstate Monica
    Commented Jul 28, 2017 at 14:21
  • 1
    $\begingroup$ Note that it does not make sense to compare a GLMM & a GEE model, as they are answering different questions. It may help to read this. $\endgroup$
    – gung - Reinstate Monica
    Commented Jul 28, 2017 at 14:23

1 Answer 1

Reset to default
2
$\begingroup$

More of a too long comment.

I really want to study the effect of temperature and precipitation, and I am thinking of using them as fixed factors, but it does not sound correct because they are a range of values. I have thought about using the values as an offset in the model, ...

Offsets are mostly used for modeling rates, and temperature/precipitation are not involved in rates, so modeling them using offsets does not seem right. Apart from this, that temperature information is max/min-values, is a separate problem, but it is not solved by offsets! Maybe use max/min-temp as separete predictors?

Improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.