ctPostPredict {ctsemOMX} | R Documentation |
Posterior predictive type check for ctsemFit.
Description
Samples data according to the ctsemFit object, computes quantiles over time based on model fit, plots these against original data.
Usage
ctPostPredict(
fit,
timestep = 0.1,
n.subjects = 100,
probs = c(0.025, 0.5, 0.975),
plot = TRUE,
ctPlotArrayArgs = list(grid = FALSE, legend = FALSE),
indPlotArgs = list(colourby = "subject", lwd = 2, new = FALSE, type = "p", opacity =
0.3),
mfrow = "auto"
)
Arguments
fit |
object of class ctsemFit as returned from |
timestep |
positive value denoting the time interval to use for sampling. |
n.subjects |
Number of subjects worth of data to sample. |
probs |
Vector of values between 0 and 1 denoting quantiles to generate. For plotting, vector should be of length 3 and values should be rising. |
plot |
Whether to plot or return the generated data. |
ctPlotArrayArgs |
additional arguments to pass to |
indPlotArgs |
list of parameters to pass to ctIndplot, for plotting original data. Only used if plot=TRUE. |
mfrow |
2 dimensional integer vector defining number of rows and columns of plots,
as per the mfrow argument to |
Value
Either nothing (if plot=TRUE) or an array containing generated data over quantiles.
Examples
data("AnomAuth")
AnomAuthmodel <- ctModel(LAMBDA = matrix(c(1, 0, 0, 1), nrow = 2, ncol = 2),
Tpoints = 5, n.latent = 2, n.manifest = 2, MANIFESTVAR=diag(0, 2), TRAITVAR = 'auto')
AnomAuthFit <- ctFit(AnomAuth, AnomAuthmodel)
ctPostPredict(AnomAuthFit,timestep=.5,n.subjects=100)