Generate Constraint Matrices

Generate a constraint matrix with a predefined structure

constrMat(n, type = c("monotone", "control", "average", "laverage",
  "uaverage", "caverage"), base = 1)

Arguments

n	a (possibly named) vector of sample sizes for each group
type	character string defining the type of constraints; one of "monotone", "control","average","laverage","uaverage", or "caverage"
base	column of the constraint matrix representing a control group (when type = "control")

Value

a constraint matrix

See also

orlm, constrSet

Examples

n <- c(10,20,30,40)
constrMat(n, type="monotone")
#>       1  2  3 4
#> [1,] -1  1  0 0
#> [2,]  0 -1  1 0
#> [3,]  0  0 -1 1
constrMat(n, type="control", base=2)
#>      1  2 3 4
#> [1,] 1 -1 0 0
#> [2,] 0 -1 1 0
#> [3,] 0 -1 0 1
constrMat(n, type="average")
#>               1          2          3         4
#> [1,] -1.0000000  0.2222222  0.3333333 0.4444444
#> [2,] -0.3333333 -0.6666667  0.4285714 0.5714286
#> [3,] -0.1666667 -0.3333333 -0.5000000 1.0000000
constrMat(n, type="laverage")
#>               1          2    3 4
#> [1,] -1.0000000  1.0000000  0.0 0
#> [2,] -0.3333333 -0.6666667  1.0 0
#> [3,] -0.1666667 -0.3333333 -0.5 1
constrMat(n, type="uaverage")
#>       1          2          3         4
#> [1,] -1  0.2222222  0.3333333 0.4444444
#> [2,]  0 -1.0000000  0.4285714 0.5714286
#> [3,]  0  0.0000000 -1.0000000 1.0000000
constrMat(n, type="caverage", base=2)
#>          1  2     3   4
#> [1,] 1.000 -1 0.000 0.0
#> [2,] 0.250 -1 0.750 0.0
#> [3,] 0.125 -1 0.375 0.5