3 \title{Shimodaira-Hasegawa Test}
5 sh.test(..., x, model = DNAmodel(), B = 100)
8 \item{...}{either a series of objects of class \code{"phylo"}
9 separated by commas, or a list containing such objects.}
10 \item{x}{a list, a matrix, or a data frame containing the (aligned)
12 \item{model}{the model to be fitted to each tree (as an object of
14 \item{B}{the number of bootstrap replicates.}
17 This function computes the Shimodaira--Hasegawa test for a set of
21 The present implementation follows the original formulation of
22 Shimodaira and Hasegawa (1999). A difference is that the bootstrap
23 resampling is done on the original sequence data rather than the RELL
24 method as sugested by Shimodaira and Hasegawa.
27 a numeric vector with the P-value associated with each tree given in
31 Shimodaira, H. and Hasegawa, M. (1999) Multiple comparisons of
32 log-likelihoods with applications to phylogenetic
33 inference. \emph{Molecular Biology and Evolution}, \bold{16},
36 \author{Emmanuel Paradis \email{Emmanuel.Paradis@mpl.ird.fr}}
38 \code{\link{mlphylo}}, \code{\link{DNAmodel}}
42 t1 <- nj(dist.dna(woodmouse))
43 t2 <- rtree(15, tip.label = t1$tip.label)
44 t3 <- rtree(15, tip.label = t1$tip.label)
45 ### Are the NJ tree and two random tress significantly different?
46 \dontrun{sh.test(t1, t2, t3, x = woodmouse, B = 100)}