umx

Posted on
No user picture. qingwen Joined: 08/15/2019

How to add CIs in umx?

Forums

I made a common pathway model in umx. I add CIs through adding umxCI(model, run="yes"). However, the output are NAs and there is a warning that "3: The nonlinear constraints and bounds could not be satisfied. The problem may have no feasible solution." I wonder what it means and what I can do to deal with it.

Posted on
No user picture. juan_q Joined: 06/17/2019
Posted on
No user picture. Liz Joined: 04/20/2016

How to summarize umx results?

Forums

Dear umx experts,

I am new to umx. May I ask how could I get the a2,c2,e2 and their CIs from the results of umxACEv, as I did with OpenMx? And how could I calculated the genetic correlation between two phenotypes with umx? I tried the demo data and found that the summary of the estimated model only reported a1,c1,e1,a2,c2,e2, and the c in umxACEv is negative. I also tried a univariate ACE model. The cl was blank. May I ask how could I use these information to summary the results which are generally reported? Any kind suggestions would be appreciated.

Posted on
Picture of user. tbates Joined: 07/31/2009

mxTryHard and loglikelihood differs from other runs of a model

Forums

A user asks
> When we tried to run the tutorial script from your preprint PeerJ paper (Bates, Maes and Neale, 2017), our script doesn't produce the same overall estimate or standardised estimates as in the PeerJ paper (e.g. we have 27,287.23 as the `-2*loglikelihood` instead of 12,186.28 in the PeerJ paper).

Generally, if you’re in any doubt about a model solution or if there’s a code or warning after the fun, it's worth trying:

```Splus
m1 = mxTryHardOrdinal(m1)
```

Posted on
Picture of user. tbates Joined: 07/31/2009

umxVersion() vs. umxVersion(“umx”).

Forums

"How much of a problem is this error when we run umXVersion(“umx”)?:
"Error in mxOption(model, "Default optimizer") : trying to get slot "options" from an object of a basic class ("character") with no slot?”

You should just run
umxVersion()

It is thinking that the first item you provide is a model, and tries to see what version you ran under.

Posted on
Picture of user. tbates Joined: 07/31/2009

How do I distinguish paths by group in multiple group models?

Forums

By default, `umxRAM` labels paths using s simple "A_to_B" style.
To distinguish what group a path is in, simply add `suffix = ` to the `uxmRAM` call. This will add the suffix string to each path. (for one-off differences, just add the label manually inside `umxPath`).

so

```
m1 = umxRAM("firstGroup", data=data, suffix = "_group1",
umxPath(v.m. = c("A", "B")),
umxPath('A', with = 'B')
)
```
Would contain paths like "A_with_B_group1"

Posted on
Picture of user. tbates Joined: 07/31/2009
Posted on
Picture of user. tbates Joined: 07/31/2009

How does umx handle categorical data?

Forums

How does your umx handle categorical data? Does it handle ordinal and unordered categories in the same way and what estimator is it using?

Posted on
Picture of user. tbates Joined: 07/31/2009

tutorials on umx

Forums

There is a growing set of tutorials on using umx here

http://tbates.github.io

There's also a paper under review at Journal of Statistical Software, and you can get a pre-print here:

https://www.dropbox.com/s/fhyj8ps2cnfzu6m/bates.pdf?dl=1

Posted on
Picture of user. tbates Joined: 07/31/2009

installing umx

Forums

you can install umx as with other packages on CRAN


install.packages("umx")
library("umx")

If you want the bleeding-edge version, install from github


devtools::install_github("tbates/umx")
library("umx")