Login

News

Difference between optimizers

Posted on
No user picture. ReddyM Joined: 10/02/2019
Forums
Behavioral Genetics Models
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)

Replied on Thu, 10/03/2019 - 10:47
Picture of user. AdminRobK Joined: 01/24/2014

Should i be concerned by this, or just use slsqp for the cases csolnp fails?

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.

Is there a particular reason for this?

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
  Login

News