ctab {gRain} | R Documentation |
General representation of multidimensional
tables (by ctab
objects).
Operations on these include multiplication, division and
marginalization.
ctab(varNames, nLevels, values = 1, normalize = c("none", "first", "all"), smooth = 0) ctabop(t1, t2, op="*") ctabmarg(t1, marg, normalize=FALSE) ## S3 method for class 'ptab': varNames(x) ## S3 method for class 'ptab': nLevels(x) ## S3 method for class 'ptab': valueLabels(x)
varNames |
Names of variables defining table |
nLevels |
Levels of variables |
values |
Table values |
t1,t2,x |
Objects of class "ptab" |
op |
Operation to perform. Either "*" or "/" (where 0/0=0) |
marg |
Set of nodes to marginalize onto |
normalize |
Should result be normalized, see 'Details' below. |
smooth |
Should values be smoothed, see 'Details' below. |
A ctab object represents a table defined by a set of variables and their levels, together with the values of the table. E.g. f(a,b,c) can be a table with a,b,c representing levels of binary variable
If normalize="first"
then for each configuration of the parents,
"pa", the probabilities are normalized to sum to one. Thus f(a,b,c)
becomes a conditional probability table of the form p(a|b,c). If
normalize="all"
then the sum over all entries of f(a,b,c) is
one.
If smooth
is non–zero then zero entries of
values
are replaced with smooth
before normalization takes place.
All ctab functions return an object of class ctab
.
Søren Højsgaard, sorenh@agrsci.dk
t1 <- ctab(c("gender","answer"),list(c('male','female'),c('yes','no')), values=1:4) t2 <- ctab(c("answer","category"), list(c('yes','no'),c(1,2)), values=1:4+10) t3 <- ctab(c("category","foo"), c(2,2), values=1:4+100) varNames(t1) nLevels(t1) valueLabels(t1) ctabop(t1,t2, "*") ctabop(t1,t2, "/") tgender <- ctabmarg(t1, "gender") t1 tgender ctabop(t1,tgender,"/")