scrime (version 1.3.5)

rowTrendStats: Rowwise Linear Trend Tests

Description

Computes for each row of a matrix the value of the statistic of a linear trend test for testing whether the ordinal variable corresponding to the row of the matrix is associated with an ordinal response.

In the two-class case, the statistic of the Cochran-Armitage trend test is computed by default.

Usage

rowTrendStats(X, y, use.n = NULL, add.pval = TRUE)

Arguments

X

a numeric matrix in which each row represents an ordinal variable and each column corresponds to an observation. The entries of this matrix are interpreted as scores for the different (ordered) levels of the respective variables.

y

a numeric vector of length ncol(X) containing the class labels of the observations represented by the columns of X, where these labels are interpreted as scores for the different classes.

use.n

should the squared values of Pearson's correlation coefficient be multiplied by ncol(X) to generate the values of the test statistic? If FALSE, the squared values are multiplied by ncol(X) - 1. By default, the squared values are multiplied by ncol(X) if y shows two levels, leading to the Cochran-Armitage test of trend. Otherwise, they are multiplied by ncol(X) - 1 leading to the linear trend test statistic of Mantel (1963) described, e.g., on page 87 of Agresti (2002).

add.pval

should p-values be added to the output? If FALSE, only the rowwise values of the linear trend test statistic will be returned. If TRUE, additionally the (raw) p-values based on an approximation to the ChiSquare-distribution with 1 degree of freedom are returned.

References

Agresti, A.\ (2002). Categorical Data Analysis. Wiley, Hoboken, NJ. 2nd Edition.

Armitage, P.\ (1955). Tests for Linear Trends in Proportions and Frequencies. Biometrics, 11, 375-386.

Cochran, W.~G.\ (1954). Some Methods for Strengthening the Common ChiSquare Tests. Biometrics, 10, 417-451.

Mantel, N.\ (1963). Chi-Square Test with one Degree of Freedom: Extensions of the Mantel-Haenszel Procedure. Journal of the American Statistical Association, 58, 690-700.

See Also

rowMsquares, rowCATTs, rowChisqMultiClass

Examples

# NOT RUN {
# Generate a matrix containing data for 10 categorical 
# variables with levels 1, 2, 3.

mat <- matrix(sample(3, 500, TRUE), 10)

# Now assume that the first 25 columns correspond to
# cases and the remaining 25 columns to cases. Then
# a vector containing the class labels is given by

cl <- rep(0:1, e=25)

# The values of the Cochran-Armitage trend test can
# then be computed by

rowTrendStats(mat, cl)

# This leads to the same results as

cases <- rowTables(mat[, cl==1])
controls <- rowTables(mat[,cl==0])

rowCATTs(cases, controls)

# or as

out <- rowMsquares(cases, controls)
n <- ncol(mat)
out$stats * n / (n - 1)

  
# }