is called MVPARTwrap and the function is MRT(). Its output comprises two graphs
and a lot of numerical results. Let us apply it to our previous result.
# Extracting MRT results from an mvpart object
spe.ch.mvpart.wrap summary(spe.ch.mvpart.wrap)
Hint Argument percent indicates the smallest percentage of variance explained by a
species at a node that one considers interesting. No test is made here; it is an
arbitrarily chosen value.
Cogo
Satr
Phph
Babl
Thth
Teso
Pato
Lele
Sqce
Baba
Albi
Gogo
Eslu
Pefl
Legi
Cyca
Titi
Ruru
leaf # 3
Cogo
Satr
Phph
Babl
Thth
Teso
Chna
Pato
Lele
Sqce
Baba
Albi
Gogo
Eslu
Pefl
Rham
Legi
Scer
Cyca
Titi
Ruru
Alal
Anan
leaf # 4
Chna
Lele
Sqce
Gogo
Eslu
Legi
Scer
Gyce
Ruru
Blbj
Alal
leaf # 6
Cogo
Satr
Phph
Babl
Thth
Teso
Chna
Pato
Lele
Sqce
Baba
Albi
Gogo
Eslu
Pefl
Rham
Legi
Scer
Cyca
Titi
Abbr
Icme
Gyce
Ruru
Blbj
Alal
Anan
leaf # 7
Fig. 4.28 Pie chart of the fish composition of the four leaves retained in the MRT analysis
4.12 Multivariate Regression Trees (MRT): Constrained Clustering
133