stm (version 1.3.5)

plot.STMpermute: Plot an STM permutation test.

Description

Plots the results of a permutation test run using permutationTest.

Usage

# S3 method for STMpermute
plot(x, topic, type = c("match", "largest"),
  xlim = NULL, ylim = NULL, ...)

Arguments

x

Object from the output of permutationTest.

topic

Integer indicating which topic to plot.

type

Character string indicating what topic comparison to use. "match" uses the Hungarian aligned method and "largest" uses the largest mean in direction of reference topic.

xlim

Range of the X-axis.

ylim

Range of the Y-axis.

...

Other parameters which may be passed to plot.

Details

This function plots the output of permutationTest by stacking horizontal confidence intervals for the effects of the permuted variable. In choosing the topic in the permuted runs of stm to plot the effect for, two methods are available, "match" and "largest". The former uses Kuhn's (1955) Hungarian method to align the topics, and then uses the model's best match of the reference topic. The latter uses the topic which has the expected effect size in the direction of the reference model effect; thus, we would expect this method to be quite conservative.

See Also

permutationTest

Examples

# NOT RUN {
# }
# NOT RUN {
temp<-textProcessor(documents=gadarian$open.ended.response,metadata=gadarian)
out <- prepDocuments(temp$documents, temp$vocab, temp$meta)
documents <- out$documents
vocab <- out$vocab
meta <- out$meta
set.seed(02138)
mod.out <- stm(documents, vocab, 3, prevalence=~treatment + s(pid_rep), data=meta)
summary(mod.out)
prep <- estimateEffect(1:3 ~ treatment + s(pid_rep), mod.out, meta)
plot(prep, "treatment", model=mod.out,
     method="difference",cov.value1=1,cov.value2=0)
test <- permutationTest(formula=~ treatment + s(pid_rep), stmobj=mod.out, 
                        treatment="treatment", nruns=25, documents=documents,
                        vocab=vocab,data=meta, stmverbose=FALSE)
plot(test,2, xlab="Effect", ylab="Model Index", main="Topic 2 Placebo Test")
# }