Sometimes, plot.sienaGOF gives incorrect positions for the (red) observed values relative the violins indicating the distributions of simulated values.
An example script is
library(RSiena)
network <- as_dependent_rsiena(array(c(s501,s502), dim = c(50, 50, 2 ) ) )
mydata <- make_data_rsiena(network)
myeff <- make_specification(mydata)
alg_alg <- set_algorithm_saom(seed=1, nsub=0)
ans <- siena(data=mydata, effects=myeff, returnDeps=TRUE, control_algo=alg_alg)
gof <- test_gof(ans, OutdegreeDistribution, varName="network", cumulative=FALSE, levls=1:10)
plot(gof)
gofi <- test_gof(ans, IndegreeDistribution, varName="network", cumulative=FALSE, levls=1:10)
plot(gofi)
When the levls=1:10 argument is omitted, the error does not occur.
For gofi, even the ordering of the values is incorrect.
The two plots are attached.
Provide system information
- OS: Windows.
- R version: 4.5.1
- autograph version: 0.6.0
- manynet version: 1.7.2.
Sometimes, plot.sienaGOF gives incorrect positions for the (red) observed values relative the violins indicating the distributions of simulated values.
An example script is
library(RSiena)
network <- as_dependent_rsiena(array(c(s501,s502), dim = c(50, 50, 2 ) ) )
mydata <- make_data_rsiena(network)
myeff <- make_specification(mydata)
alg_alg <- set_algorithm_saom(seed=1, nsub=0)
ans <- siena(data=mydata, effects=myeff, returnDeps=TRUE, control_algo=alg_alg)
gof <- test_gof(ans, OutdegreeDistribution, varName="network", cumulative=FALSE, levls=1:10)
plot(gof)
gofi <- test_gof(ans, IndegreeDistribution, varName="network", cumulative=FALSE, levls=1:10)
plot(gofi)
When the levls=1:10 argument is omitted, the error does not occur.
For gofi, even the ordering of the values is incorrect.
The two plots are attached.
Provide system information