Hi everyone!
I have data consisting of several response variables and several explanatory
variables. I wish to do a permanova on this using the vegan library and the
adonis() function. However, my data had several missing values in it. In
order to 'fix' this I used the mice() function from the mice library to
make
5 imputations for all the missing values. To do analysis on the 5 datasets
with different imputed values, the mice library includes the with.mids()
function in which you can call the mice object and calculate a model for all
the 5 datasets. The results from these analysis can later on be pooled with
the pool() function from the mice library.
The problem I have is that I am not sure how to include a permanova (with
the adonis() function) as an expression in the with.mids() function.
In order to use the mice() function and calculate multiple imputations I
needed one dataframe with all of my response and explanatory variables. In
order to use the adonis() function (I think) I need a separate dataframe
with my response variables and a separate dataframe with my explanatory
variables.
Does anyone know how to use the output from a multiple imputation by chained
equations (mice) to do a permanova?
I hope my question is clear, and if anyone has any ideas or suggestions,
your help would be greatly appreciated.
Thanks!
--
View this message in context:
http://r.789695.n4.nabble.com/permanova-on-MICE-object-tp2318342p2318342.html
Sent from the R help mailing list archive at Nabble.com.
