| JoinModels {cmm} | R Documentation |
Returns the simultaneous specification of two models
JoinModels(...)... |
list of ‘compatible’ models, i.e., each model is specified using the same number of functions (and matrices) |
Models can be of any form allowed in CMM (see MarginalModelFit), eg, list(bt,coeff,at), with the restriction that the number of columns of the
at matrices must be equal, and the list of functions in coeff must be identical.
CMM model form
W. P. Bergsma w.p.bergsma@lse.ac.uk
Bergsma, W. P. (1997). Marginal models for categorical data. Tilburg, The Netherlands: Tilburg University Press. http://stats.lse.ac.uk/bergsma/pdf/bergsma_phdthesis.pdf
Bergsma, W. P., Croon, M. A., & Hagenaars, J. A. P. (2009). Marginal models for dependent, clustered, and longitudunal categorical data. Berlin: Springer.
DirectSum, SpecifyCoefficient, MarginalModelFit
# simultaneously specify marginal independence in two marginal tables
bt1 = ConstraintMatrix(c("A","B"),list(c("A"),c("B")),c(3,3))
at1 = MarginalMatrix(c("A","B","C"),c("A","B"),c(3,3,3))
model1 = list(bt1,"log",at1)
bt2 = ConstraintMatrix(c("B","C"),list(c("B"),c("C")),c(3,3))
at2 = MarginalMatrix(c("A","B","C"),c("B","C"),c(3,3,3))
model2 = list(bt2,"log",at2)
model12 = JoinModels(model1,model2)
# the model can be fitted to an artificial data set
n = c(1:27)
fit=MarginalModelFit(n,model12)