| twostagechao {vegan} | R Documentation |
The function implements the two-stage probabilistic approach to multiple-community similarity indices proposed by Chao et al. (2008), extending the traditional pairwise comparisons of sites.
twostagechao(x, order = 2, N = nrow(x), m = 1, nboot = 200, subset)
x |
A community matrix with rows as sites and columns as species. |
order |
The order of the similarity, see details. |
N |
The number of sites to use. By default it is the number of rows in x.
If N < ncol(x), then N and subset must conform
(length(subset) == N).
|
m |
The exponent used for calculating the NESS index variant, see details. |
nboot |
Number of bootstrap samples to use for calculating standard error of the estimate.
If nboot = 0, standard error is not computed.
|
subset |
Optional vector defining the subset of sites to use in the calculations.
If this argument is provided, its length should equal N.
|
PUT DETAILS HERE
The result is an object of class 'twostagechao' inheriting its structure from class 'htest'.
P\'eter S\'olymos, solymos@ualberta.ca
Chao, A., Jost, L., Chiang, S. C., Jiang, Y.-H., Chazdon, R. L. (2008) A two-stage probabilistic approach to multiple-community similarity indices. Biometrics 64, 1178–1186.
Supplementary material at: http://www.biometrics.tibs.org/datasets/070539.pdf
data(BCI)
twostagechao(BCI)
res1 <- t(sapply(2:5, function(z) twostagechao(BCI, order=z)$statistic))
rownames(res1) <- paste("Order =", 2:5)
## similarity decreases as rare species get more weights
res1
## original example from Chao et al. 2008
## supplementary material, LSUR Primary Plot
lep1 <- structure(c(7L, 48L, 17L, 0L, 38L, 14L, 31L, 37L, 121L, 6L, 30L,
16L, 0L, 27L, 10L, 0L, 24L, 5L, 2L, 23L, 6L, 3L, 21L, 6L, 1L,
19L, 1L, 0L, 19L, 73L, 0L, 19L, 2L, 2L, 17L, 4L, 3L, 17L, 17L,
2L, 16L, 5L, 0L, 16L, 11L, 0L, 16L, 4L, 2L, 15L, 1L, 0L, 15L,
9L, 4L, 14L, 7L, 0L, 14L, 2L, 0L, 13L, 4L, 5L, 12L, 0L, 1L, 11L,
7L, 2L, 11L, 20L, 1L, 9L, 0L, 2L, 9L, 5L, 0L, 9L, 0L, 1L, 8L,
0L, 0L, 8L, 6L, 0L, 8L, 0L, 0L, 8L, 0L, 0L, 8L, 0L, 3L, 7L, 3L,
0L, 7L, 17L, 0L, 7L, 0L, 1L, 0L, 0L, 1L, 0L, 0L, 1L, 0L, 0L,
1L, 0L, 0L, 1L, 0L, 0L, 2L, 6L, 1L, 0L, 6L, 6L, 0L, 6L, 2L, 1L,
5L, 0L, 2L, 5L, 1L, 1L, 5L, 3L, 0L, 5L, 2L, 0L, 5L, 1L, 0L, 5L,
1L, 1L, 4L, 1L, 1L, 4L, 2L, 0L, 4L, 29L, 0L, 4L, 3L, 0L, 4L,
1L, 0L, 4L, 0L, 0L, 3L, 2L, 0L, 3L, 1L, 0L, 3L, 0L, 3L, 2L, 0L,
1L, 2L, 0L, 1L, 2L, 0L, 1L, 2L, 3L, 0L, 2L, 7L, 0L, 2L, 5L, 0L,
2L, 1L, 0L, 2L, 1L, 0L, 2L, 1L, 0L, 2L, 0L, 0L, 2L, 0L, 0L, 2L,
0L, 0L, 2L, 0L, 0L, 2L, 0L, 0L, 2L, 0L, 0L, 2L, 0L, 0L, 2L, 0L,
3L, 0L, 1L, 1L, 0L, 1L, 1L, 0L, 3L, 0L, 0L, 54L, 0L, 0L, 3L,
0L, 2L, 0L, 0L, 2L, 0L, 1L, 1L, 0L, 1L, 1L, 0L, 1L, 1L, 1L, 1L,
1L, 2L, 0L, 1L, 5L, 0L, 1L, 3L, 0L, 1L, 3L, 0L, 1L, 1L, 0L, 1L,
1L, 0L, 1L, 1L, 0L, 1L, 1L, 0L, 1L, 1L, 0L, 1L, 1L, 0L, 1L, 0L,
0L, 1L, 0L, 0L, 1L, 0L, 0L, 1L, 0L, 0L, 1L, 0L, 0L, 1L, 0L, 0L,
1L, 0L, 0L, 1L, 0L, 0L, 1L, 0L, 0L, 1L, 0L, 0L, 1L, 0L, 0L, 1L,
0L, 0L, 1L, 0L, 0L, 1L, 0L, 0L, 1L, 0L, 0L, 1L, 0L, 0L, 1L, 0L,
2L, 0L, 0L, 2L, 0L, 0L, 1L, 0L, 0L, 0L, 0L, 3L, 0L, 0L, 2L, 0L,
0L, 1L, 0L, 0L, 1L, 0L, 0L, 1L), .Dim = c(3L, 120L), .Dimnames = list(
c("Tree", "Sapling", "Seedling"), NULL))
## compare values in Table 3 of Chao et al. 2008
## (4th column, last 6 rows)
## results are different!
Order <- rep(2:3, 3)
m <- c(1, 1, 5, 5, 10, 10)
res2 <- t(sapply(1:6, function(z)
twostagechao(lep1, order=Order[z], m=m[z])$statistic))
rownames(res2) <- c("Morisita C23", "Morisita C33", "NESS23(5)",
"NESS33(5)", "NESS23(10)", "NESS33(10)")
res2