[R] lost in vegan package

eva epitta at upatras.gr
Wed Sep 1 14:57:28 CEST 2010


Eva,

The printed output can be dumped to a file using command sink(). See its
help
page for info.

The oecosimu result is the output of the basic statistic function amended
with
item "oecosimu" with structure:

> str(mod$oecosimu)
List of 6
 $ z          : num -66.5
 $ pval       : num 0.01
 $ simulated  : num [1, 1:99] 159621 160463 161865 161720 162472 ...
 $ method     : chr "r00"
 $ statistic  : Named num 54904
  ..- attr(*, "names")= chr "statistic"
 $ alternative: chr "two.sided"

You can list and save itmes like mod$oecosimu$statistic, mod$oecosimu$z,
mod$oecosimu$pval which all are combined into printed ouput. You can use
commands like write.table, write.csv to save these results for external
software.

Please note the 'alternative' algorithm: it seems to me that your hypotheses
may
one-sided.

Cheers, Jari Oksanen 

Dr Oksanen

I'm checking if the simulated number of checkerboard units of a species pair
is larger or smaller than the observed number of checkerboard units in the
original dataset. If it is larger I list the species pair as exhibiting
cooccurrence and if it is smaller i list it as exhibiting competition. The
procedure I'm following (which I'm not sure is correct) is that I use the
probabilities i get from the output at the 0.1 level. 
As I understand, this means that I'm checking at the 0,05 level in each tail
of the distribution. Am I wrong?

Thank's again
Eva


-- 
View this message in context: http://r.789695.n4.nabble.com/lost-in-vegan-package-tp2400145p2403096.html
Sent from the R help mailing list archive at Nabble.com.



More information about the R-help mailing list