Description Usage Arguments Details Value Author(s) References Examples
View source: R/cumres.lvmfit.R
Calculates GoF statistics based on cumulative residual processes for
structural equation models fitted with the lava
package.
1 2 3 4 5 6 7 8 9 10 11 12 13 
model 

y 
A formula specifying the association to be checked. Alternatively the outcome specified as a function or a string with the name of the outcome in the model. 
x 
Predictor. A function, vector or character 
full 
If FALSE the prediction, Pr, of the variable that are ordered after is only calculated based on the conditional distribution given covariates. If TRUE the conditional expectation is based on the largest set of covariates and endogenous variables such that the residual and Pr are uncorrelated. 
data 
data.frame (default is the model.frame of the model) 
p 
Optional parameter vector 
R 
Number of processes to simulate 
b 
Moving average parameter 
plots 
Number of processes to save for use with the plot method 
... 
Additional arguments parsed on to lowerlevel functions 
With y
and x
given as functions the user can decide which
variables to use in the prediction of the outcome and predictor (use the
predict
method as below).
Returns a cumres
object with associated
plot
,print
,coef
methods
Klaus K. Holst
B.N. Sanchez and E. A. Houseman and L. M. Ryan (2009) ResidualBased Diagnostics for Structural Equation Models. Biometrics Volume 65 (1), pp 104115.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16  library(lava)
m < lvm(list(c(y1,y2,y3)~eta,eta~x)); latent(m) < ~eta
## simulate some data with nonlinear covariate effect
functional(m,eta~x) < function(x) 0.3*x^2
d < sim(m,100)
e < estimate(m,d)
## Checking the functional form of eta on x
g < cumres(e,eta~x,R=1000)
plot(g)
x < function(p) predict(e,x=~y2+y3,p=p)[,"eta"]
## Checking the functional form of y1 on eta
cumres(e,y1~eta,R=1000)
g < cumres(e,"y1",x=x,R=1000)
plot(g)

Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.