OpenSEM Forums
noamm
Joined: 10/10/2018
ReddyM
Joined: 10/02/2019
mxbootstrap
Hi,
Is there a way to write out to console the number of replication being computed? Just because for some models it can take a lot of time to bootstrap, and this way would be easier to know when it will be finished.
Thanks in advance, ReddyM
- Read more about mxbootstrap
- 2 comments
- Log in or register to post comments
George Richardson
Joined: 10/18/2018
Inconsistency between global chi-square difference tests and t-tests
Hi,
I've specified a multivariate ACE model and found that the z/t-tests produce substantially different p-values than the chi-square difference test per parameter constrained to zero. My recollection is that these tests should produce equivalent results but I recall reading somewhere that at relatively small n (here, n = 260 pairs), the global test may have more power. Does anyone know if this is likely correct or if there are methods papers address it? Could the difference be due to correlations among the z/t-tests? Not sure what's going on here.
JuanJMV
Joined: 07/20/2016
Problem to get the genetic correlations
Hi everyone,
I have fitted a 4-variate model and I have fixed A to 0 for one of these phenotypes. The model fits OK but I am struggled to obtain the genetic correlations.
This is the output from the model but as one of the A parameters is fixed to 0 I get these results:
- Read more about Problem to get the genetic correlations
- 8 comments
- Log in or register to post comments
liagym1
Joined: 11/06/2019
ACE with moderator- warning in output
I'm running an ACE with moderator script, on Mx. When I drop one of the parameters (U), I receive this message-
Mx starting optimization; number of parameters = 7
*** WARNING! ***
I am not sure I have found a solution that satisfies
Kuhn-Tucker conditions for a minimum.
NAG's IFAIL parameter is 6
Looks like I got stuck here. Check the following:
1. The model is correctly specified
2. Starting values are good
3. You are not already at the solution
- Read more about ACE with moderator- warning in output
- 1 comment
- Log in or register to post comments
yaning
Joined: 04/12/2019
questions about the GxE model
Hi, everyone
Some questions about the GxE model:
From Wikipedia,
- Read more about questions about the GxE model
- 6 comments
- Log in or register to post comments
lemire
Joined: 12/13/2019
Genetic correlation: two ways, two different estimates
Hi,
I implemented two ways of calculating the genetic correlation between two phenotypes, estimated from twins (direct model of rg between A1 and A2, and Cholesky).
My two implementations are based on the path diagrams from H.H. Maes' slides 18 and 53 found at
https://vipbg.vcu.edu/media/course/HGEN619_2015/BivariateGeneticAnalysis2.pdf
I do not get the same estimates between the two methods (-0.26 vs 0.51). Below is the code with the reproducible fake data and results. The code is also attached.
khusmann
Joined: 08/11/2019
Sample weights / complex study designs
Hi all,
I'm wondering what is the proper way to do sample weights in OpenMx >2.0? I found this [old thread](https://openmx.ssri.psu.edu/thread/3661) with an example that uses `mxFIMLObjective`, but `mxFIMLObjective` looks to be deprecated. Any help with a working example would be much appreciated!
Also, this way of doing weights by nesting mxModels seems pretty cumbersome -- is there an updated interface or a way to "tack on" weights to an existing model definition? (Ideally I'd like a way for models to be easy to run with and without weights).
- Read more about Sample weights / complex study designs
- 1 comment
- Log in or register to post comments
ReddyM
Joined: 10/02/2019
Ordered quantile normalization
Can i use ordered quantile normalization, to normalize a dataset for univariate ACE model? I think not, but if i could, that would be really good.
Thanks in advance!
- Read more about Ordered quantile normalization
- 2 comments
- Log in or register to post comments
diana
Joined: 04/17/2019
How to fix the genetic correlation larger than 1
Hi everyone, I am fitting a bivariate ACE model and I am very confused about the Ra (genetic correlation of two traits) that is larger than 1. Since Ra is a correlation coefficient, it should range from -1 to 1, am I right? I don't know how to explain this result.
In the ACE model, Ra=1.37 ,Rc=-0.04, Re=0.22.
Then I fitted ADE model, Ra=0.07, Rd=NA, Re=0.22.
The results of the goodness-of-fit test between saturated model and ACE or ADE are as follows.