Difference between optimizers
ReddyM
Joined: 10/02/2019
In my twin model i calculate Rmz and Rdz from the saturated model, however when i run it (i run it on several variables) for a few variables CSOLNP fails to calculate confidence intervals, while SLSQP does caculate all. Is there a particular reason for this? Should i be concerned by this, or just use slsqp for the cases csolnp fails?
Here is the sat part, if it helps:
#sat model, rmz, rdz:
PathMZ <- mxMatrix("Lower",nrow = 2,ncol = 2,free = TRUE,values=starting_value_pathcoeff,name = "PathMZ")#cholesky for pos definitness
PathDZ <- mxMatrix("Lower",nrow = 2,ncol = 2,free = TRUE,values=starting_value_pathcoeff,name = "PathDZ")#cholesky for pos definitness
expCovMZ <- mxAlgebra(expression = PathMZ%*%t(PathMZ),name = "expCovMZ")
expCovDZ <- mxAlgebra(expression = PathDZ%*%t(PathDZ),name = "expCovDZ")
Rmz <- mxAlgebra(expression = sqrt(vec2diag(1/diag2vec(expCovMZ)))%*%expCovMZ%*%sqrt(vec2diag(1/diag2vec(expCovMZ))), name = "Rmz")
Rdz <- mxAlgebra(expression = sqrt(vec2diag(1/diag2vec(expCovDZ)))%*%expCovDZ%*%sqrt(vec2diag(1/diag2vec(expCovDZ))), name = "Rdz")
expMZ <- mxExpectationNormal(covariance = "expCovMZ",means = "exp_mean_mz",dimnames = selectVariables)
expDZ <- mxExpectationNormal(covariance = "expCovDZ",means = "exp_mean_dz",dimnames = selectVariables)
MZmodel <- mxModel(intercepts,beta,dependent_vars_mz,PathMZ,Rmz, expCovMZ,exp_mean_mz,DataMZ,expMZ,fitML ,name="MZmodel")
#MZmodel <- mxModel(MZmodel,mxCI(c("Rmz")))
DZmodel <- mxModel(intercepts,beta,dependent_vars_dz,PathDZ,Rdz, expCovDZ,exp_mean_dz,DataDZ,expDZ,fitML ,name="DZmodel")
#DZmodel <- mxModel(DZmodel,mxCI(c("Rdz")))
fitmultiMLsat <- mxFitFunctionMultigroup(c("MZmodel.fitfunction","DZmodel.fitfunction"))
SatModel <- mxModel("SAT",MZmodel,DZmodel,fitmultiMLsat)
SatModel <- mxModel(SatModel,mxCI(c("MZmodel.Rmz","DZmodel.Rdz")))
Satfinal <- mxRun(SatModel, intervals = TRUE)
sum_sat_raw <- summary(Satfinal)
probably no big deal; some hints
It's probably nothing to worry about, but it's also a good idea to use `summary(Satfinal,verbose=TRUE)` to see why the confidence limits aren't being reported.
Without going into a lot of detail, those two optimizers by default use a different internal representation of the confidence-limit search. There are also theoretical reasons why SLSQP would be expected to give better confidence intervals most of the time. If you want to use CSOLNP to fit your model, but use SLSQP to find confidence intervals, then run your MxModel with `intervals=FALSE`, and then use
omxRunCI()on your fitted MxModel (which uses SLSQP by default).Log in or register to post comments
In reply to probably no big deal; some hints by AdminRobK
Thank you!
Thank you!
I think then i will use SLSQP just for the variables where CSOLNP didn't work, because generally the intervals are a bit tighter when i use CSOLNP.
Log in or register to post comments
In reply to probably no big deal; some hints by AdminRobK
Summary
By the way, it's a diagnostic error: alpha level not reached infeasible non-linear constraint.
Log in or register to post comments