JT.test: Jonckheere-Terpstra trend test
In SAGx: Statistical Analysis of the GeneChip

Description Usage Arguments Details Value Author(s) References Examples

The test is testing for a monotone trend in terms of the class parameter. The number of times that an individual of a higher class has a higher gene expression forms a basis for the inference.

1	JT.test(data, class, labs = NULL, alternative = c("two-sided", "decreasing", "increasing"), ties = FALSE)

`data`	A matrix with genes in rows and subjects in columns
`class`	the column labels, if not an ordered fctor it will be redefined to be one.
`labs`	the labels of the categories coded by class
`alternative`	two-sided, decreasing or increasing
`ties`	Adjustment for ties

Assumes that groups are given in increasing order, if the class variable is not an ordered factor, it will be redefined to be one. The p-value is calculated through a normal approximation.

The implementation owes to suggestions posted to R list.

The definition of predictive strength appears in Flandre and O'Quigley.

an object of class JT-test, which extends the class htest, and includes the following slots

`statistic`	the observed JT statistic
`parameter`	the null hypothesis parameter, if other value than 0.
`p.value`	the p-value for the two-sided test of no trend.
`method`	Jonckheere-Terpstra
`alternative`	The relations between the levels: decreasing, increasing or two-sided
`data.name`	the name of the input data
`median1 ... mediann`	the medians for the n groups
`trend`	the rank correlation with category
`S1`	Predictive strength

Per Broberg, acknowledging input from Christopher Andrews at SUNY Buffalo

Lehmann, EH (1975) Nonparametrics: Statistical Methods Based on Ranks p. 233. Holden Day
Flandre, Philippe and O'Quigley, John, Predictive strength of Jonckheere's test for trend: an application to genotypic scores in HIV infection, Statistics in Medicine, 2007, 26, 24, 4441-4454

# Enter the data as a vector
A <- as.matrix(c(99,114,116,127,146,111, 125,143,148,157,133,139, 149, 160, 184))
# create the class labels
g <- c(rep(1,5),rep(2,5),rep(3,5))
# The groups have the medians
tapply(A, g, median)
# JT.test indicates that this trend is significant at the 5% level
JT.test(data = A, class = g, labs = c("GRP 1", "GRP 2", "GRP 3"), alternative = "two-sided")

Loading required package: multtest
Loading required package: BiocGenerics
Loading required package: parallel

Attaching package: 'BiocGenerics'

The following objects are masked from 'package:parallel':

    clusterApply, clusterApplyLB, clusterCall, clusterEvalQ,
    clusterExport, clusterMap, parApply, parCapply, parLapply,
    parLapplyLB, parRapply, parSapply, parSapplyLB

The following objects are masked from 'package:stats':

    IQR, mad, sd, var, xtabs

The following objects are masked from 'package:base':

    Filter, Find, Map, Position, Reduce, anyDuplicated, append,
    as.data.frame, cbind, colMeans, colSums, colnames, do.call,
    duplicated, eval, evalq, get, grep, grepl, intersect, is.unsorted,
    lapply, lengths, mapply, match, mget, order, paste, pmax, pmax.int,
    pmin, pmin.int, rank, rbind, rowMeans, rowSums, rownames, sapply,
    setdiff, sort, table, tapply, union, unique, unsplit, which,
    which.max, which.min

Loading required package: Biobase
Welcome to Bioconductor

    Vignettes contain introductory material; view with
    'browseVignettes()'. To cite Bioconductor, see
    'citation("Biobase")', and for packages 'citation("pkgname")'.

  1   2   3 
116 143 149 
class was not an ordered factor.  Redefined to be one.

	Jonckheere-Terpstra

data:  A by g
J = 16, p-value = 0.02311
alternative hypothesis: two-sided: GRP 1two-sided: GRP 2two-sided: GRP 3