M.fct: Marginalizing Matrix Based on Strata Information

In cta: Contingency Table Analysis Based on ML Fitting of MPH Models

Description

Creates the marginalizing matrix M using strata information found in input variable strata. That is, Mp gives the marginal probabilities corresponding to the levels of factor strata.

Usage

M.fct(strata, ncells = length(strata))

Arguments

strata	The vector that gives the stratum membership identifier.
ncells	Number of contingency table cells. Default: ncells = length(strata).

Value

M.fct returns the marginalizing matrix M.

Note

Marginals are ordered according to the levels of factor strata.

Examples:

V1	V2	y
b	yes	15
a	no	12
a	yes	13
b	yes	5
b	no	1

M1 <- M.fct(V1)
M1 %*% y

	[,1]
a	25
b	21

M2 <- M.fct(V2)
M2 %*% y

	[,1]
no	13
yes	33

M12 <- M.fct(paste(V1, V2))
M12 %*% y

		[,1]
a	no	12
a	yes	13
b	no	1
b	yes	20

Author(s)

Joseph B. Lang

References

Lang, J. B. (2004) Multinomial-Poisson homogeneous models for contingency tables, Annals of Statistics, 32, 340–383.

Examples

M.fct(rep(1, 9))
M.fct(seq(1, 9))
M.fct(c(1, 1, 2, 3, 3))

cta documentation built on Aug. 24, 2021, 1:06 a.m.