I want to model the spatial distribution of the ecological niche of cowpea, I
have a set of data (environmental variables and bioclimatic) and data of
presence of the varieties of cowpea.I wanted to use the software Maxent for do
it; But I understood through the literature that the R software is now widely
used. It is with this in mind that I would like to focus on this. By the way, I
need help to understand the procedures for doing this modeling using R. 1. I
would like to do a Shapiro-Wilks normality test, the null hypothesis of which
is to assert that The data follow a normal distribution and where the
alternative hypothesis implies an asymmetric or abnormal distribution, makes it
possible to conclude that the observations are of very asymmetric distribution.
2. I would like to do a logarithmic transformation of the response data is
therefore performed in order to reduce the asymmetry of the data and thus
improve the power of the statistical tests.3. I would like to do a canonical
redundancy analysis on all the data tables. 4. I would like to make a
bidirectional progressive selection in order to select the relevant variables
according to Akaike's information criterion or AIC and Of the
representativeness (p-value) .5. I would like to make a progressive selection
by adding, according to the adjusted R-square, in order to compare the result
of the selection 6. Finally, the procedures to be followed to model the
distribution of the varieties.
Best regards
SADDA Abou-Soufianou --------------------------------------
DoctorantUniversité Dan Dicko Dankoulodo de Maradi,120 avenue Maman Koraou,ADS,
MaradiBP 465 Maradi/Niger E-mail: a.soufianou@yahoo.frGSM : (+227)96269987
[[alternative HTML version deleted]]
_______________________________________________
R-sig-Geo mailing list
R-sig-Geo@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-geo
