General SEM Discussions
http://openmx.psyc.virginia.edu/taxonomy/term/17/0
enPrevent Calculation of Auxiliary Variable Descriptive Statistics
http://openmx.psyc.virginia.edu/thread/3127
<p>Hi All,</p>
<p>I am using OpenMx in conjunction with SEM Trees. As a result of this, I have to include the whole dataset (30,000 x 1200). Right now, it takes about 30 seconds to run the current code:</p>
<p>hpc20 <- mxModel("1 Variable",type="RAM",<br />
mxData(observed=pt_run1,type="raw"),<br />
manifestVars="A_AbsRea",<br />
mxPath(<br />
from= "A_AbsRea" ,<br />
arrows=2,free=T,values=9,labels= "e1"),<br />
mxPath(<br />
from="one",<br />
to="A_AbsRea",<br />
arrows=1,free= T,values= 8,labels = "mean1")) </p>
<p>system.time(hpc20Fit <- mxRun(hpc20,suppressWarnings=T,silent=T))<br />
user system elapsed<br />
30.723 2.092 32.806 </p>
<p>I am trying to figure out a way to speed this up. Because the computation takes so long in using SEM Trees with such a large dataset, we are starting with the simplest possible model. I suspect the model takes so long to run because of the computation of the descriptives for all of the other variables. When the dataset is changed to just include the one variable, it takes a split second. </p>
<p>I have tried to change this option thus far:</p>
<p>hpc20 <- mxOption(hpc20,'No Sort Data',"A_AbsRea")</p>
<p>although it doesn't off change the computation time. Is there a different way to have OpenMx only calculate values for the 1 variable of interest? Sorry if I am not clear or am missing something quite obvious.</p>
<p>Thanks<br />
Ross</p>
http://openmx.psyc.virginia.edu/thread/3127#commentsGeneral SEM DiscussionsTue, 27 May 2014 20:35:14 +0000jacobucc3127 at http://openmx.psyc.virginia.eduhow to make fit model?
http://openmx.psyc.virginia.edu/thread/2966
<p>hallo,<br />
I want to ask about optimalization my SEM model. i confused how to set a starting value for each variable? i send my code but in result, i get still NaN in the standar estimation.<br />
in my result, RMSEA still more than 0.08 and the CFI still less than 0.9. so, what should i do?</p>
<table id="attachments" class="sticky-enabled">
<thead><tr><th>Attachment</th><th>Size</th> </tr></thead>
<tbody>
<tr class="odd"><td><a href="http://openmx.psyc.virginia.edu/sites/default/files/OUTPUT_0.txt">OUTPUT.txt</a></td><td>9.76 KB</td> </tr>
<tr class="even"><td><a href="http://openmx.psyc.virginia.edu/sites/default/files/kodingan data Yogya Bali (DataFilter)1.txt">kodingan data Yogya Bali (DataFilter)1.txt</a></td><td>4.39 KB</td> </tr>
</tbody>
</table>
http://openmx.psyc.virginia.edu/thread/2966#commentsGeneral SEM DiscussionsWed, 30 Apr 2014 08:02:10 +0000siti nur azizah2966 at http://openmx.psyc.virginia.eduPower calculation
http://openmx.psyc.virginia.edu/thread/2959
<p>Hi.<br />
We are planning to do some moderation model analysis on the new data (which we do not have yet) and need to do power analysis before. Is there any script written for OpenMx how to do it? I found some scripts for calssical Mx, but I am not sure I am able to adjust it to moderation models and even if, the output seems to be not compatible with 64-bit Windows.<br />
I would appreciate any help for simulating data driven by such models and for calculating the power.<br />
Thank you beforehand!<br />
Julia</p>
http://openmx.psyc.virginia.edu/thread/2959#commentsGeneral SEM DiscussionsTue, 29 Apr 2014 09:20:03 +0000Julia2959 at http://openmx.psyc.virginia.eduSEM for prediction
http://openmx.psyc.virginia.edu/thread/2750
<p>Hello -</p>
<p>I wanted to check with the community if anyone has experience in using SEM for predictive modeling. If yes, how does one approach it. Thanks in advance for your help!</p>
http://openmx.psyc.virginia.edu/thread/2750#commentsGeneral SEM DiscussionsMon, 31 Mar 2014 04:07:41 +0000LearnSEM82750 at http://openmx.psyc.virginia.eduSet starting values
http://openmx.psyc.virginia.edu/thread/2712
<p>I want to ask what exactly the purpose of setting the starting value in mxPath codes? And what the basic can we use to set it? Is it like trial and error? I set some values but it always generate the NaN value in standard error, so what should I do then? Thank you.</p>
http://openmx.psyc.virginia.edu/thread/2712#commentsGeneral SEM DiscussionsTue, 25 Mar 2014 11:47:50 +0000icha2712 at http://openmx.psyc.virginia.eduFIML Estimation
http://openmx.psyc.virginia.edu/thread/2705
<p>I want to ask you.<br />
actually, I have a problem with my data. My new data is incomplete (missing data). I read on the web if no data is lost then using FIML function, but I am confused what to put where and what. I have tried but I always fail. please help me.<br />
other than that, is there any basis we determine the value? I tried to try to change the value of 0.25-1 and outcome affect the output Standard error of estimate.<br />
thank you.</p>
<table id="attachments" class="sticky-enabled">
<thead><tr><th>Attachment</th><th>Size</th> </tr></thead>
<tbody>
<tr class="odd"><td><a href="http://openmx.psyc.virginia.edu/sites/default/files/kodingan DATA BARU YogyaPabar.txt">kodingan DATA BARU YogyaPabar.txt</a></td><td>4.8 KB</td> </tr>
</tbody>
</table>
http://openmx.psyc.virginia.edu/thread/2705#commentsGeneral SEM DiscussionsTue, 25 Mar 2014 05:23:22 +0000siti nur azizah2705 at http://openmx.psyc.virginia.eduGrowth curve model with ordinal data!
http://openmx.psyc.virginia.edu/thread/2672
<p>Hi,<br />
I have 50 binary variables with values of either 0 or 1. I tried to treat these variables as ordinal and fit a growth curve model.<br />
OpenMx gave wrong message:</p>
<p>Running Linear Growth Curve Model<br />
Error: The data object 'Linear Growth Curve Model.data' contains 50 ordered factors but our ordinal integration implementation has a limit of 20 ordered factors.</p>
<p>Is that true that I only can fit 20 ordered factors using OpenMx? Could you point me a direction how to fix this problem? Any help or suggestions will be appreciated. Thanks!</p>
http://openmx.psyc.virginia.edu/thread/2672#commentsGeneral SEM DiscussionsFri, 14 Mar 2014 13:55:47 +0000Sunny Wu2672 at http://openmx.psyc.virginia.eduWhat's the relationship between a 【latent variable】and its 【indicator(s)】
http://openmx.psyc.virginia.edu/thread/2657
<p>Hi, everyone! I am a rookie of SEM.<br />
To use SEM, I have read some textbooks about it. And there are something I cannot quite sure about.</p>
<p>It is usual that a latent variable has several indicators, let's say three as an example.<br />
[indicator1] <-- (latent variable)<br />
[indicator2] <--<br />
[indicator3] <--</p>
<p>the path diagram above can be represented by equations as follow:</p>
<p>indicator1 = a1 * latent variable + error1 ... (*)<br />
indicator2 = a2 * latent variable + error2 ... (**)<br />
indicator3 = a3 * latent variable + error3 ... (***)</p>
<p>in which a1, a2, a3 means the regression weights of indicator1, 2, 3 respectively.</p>
<p>Using the equations above, I can easily replace latent variable in (*) using indicator2. </p>
<p>So my questions are:<br />
1、If indicator1 can be linearly represented by indicator2, does that mean there are linear relationship between these two indicators?<br />
2、If so. I always assume that indicators are items can be measure and represent partly of correspondent latent variable, so the more independent between indicators, the better they are. If they have some kind of linear relationship, that may cause some negative effect of the Model.</p>
<p>How can I understand these?</p>
<p>Since I am a new to SEM, my opinion may itself be wrong. So please tell me If so. </p>
<p>Thanks! :)</p>
http://openmx.psyc.virginia.edu/thread/2657#commentsGeneral SEM DiscussionsMon, 10 Mar 2014 15:21:01 +0000colin.Fu2657 at http://openmx.psyc.virginia.eduResidual variance correlations in path analysis
http://openmx.psyc.virginia.edu/thread/2573
<p>Suppose I have two endogenous variables (A and B) and I draw an arrow (double-headed) between them. What is openmx doing in the background? Is it fitting a correlation between the residuals of A and B?</p>
http://openmx.psyc.virginia.edu/thread/2573#commentsGeneral SEM DiscussionsMon, 13 Jan 2014 17:51:25 +0000fife2573 at http://openmx.psyc.virginia.edudefinition variable
http://openmx.psyc.virginia.edu/thread/2553
<p>What is a definition variable. How is it different from a moderator?</p>
http://openmx.psyc.virginia.edu/thread/2553#commentsGeneral SEM DiscussionsSat, 11 Jan 2014 07:57:36 +0000praveens12553 at http://openmx.psyc.virginia.eduComparing two models with "poor" fits?
http://openmx.psyc.virginia.edu/thread/2357
<p>I've been trying to determine the time ordering of events for some biomedical data. Suppose I have two variables (biomarker A and biomarker B), each measured at three timepoints (for a total of six variables). I have one model where biomarker A causes biomarker B. I write this model as follows:</p>
<p>mat = data.frame(matrix(c(<br />
"B1_1", "B1_2", 1, .2,<br />
"B1_2", "B1_3", 1, .2, #### biomarkers 1 cause themselves across time<br />
"B2_1", "B2_2", 1, .2,<br />
"B2_2", "B2_3", 1, .2, #### biomarkers 2 cause themselves across time<br />
"B1_1", "B2_2", 1, .2,<br />
"B1_2", "B2_3", 1, .2 #### biomarker 1 causes biomarker 2 across time<br />
), ncol=4, byrow=TRUE), stringsAsFactors=F)</p>
<p>paths = mxPath(from=mat[,1], to=mat[,2], arrows=as.numeric(mat[,3]), values=as.numeric(mat[,4]), labels=paste(mat[,1], "to", mat[,3], sep=""), free=T) </p>
<p>Then I write the opposite model (where biomarkers 2 cause biomarkers 1). When I fit both models, the RMSEA is quite good for one model and poor for the other model. However, when I look at the residual matrix, neither looks very good. (In particular, the correlations between B2_1, B2_2, and B2_3/B1_1, B1_2, and B1_3). </p>
<p>So here's my question: If I'm only interested in determining which causal relationship is better supported by the data, should I even care that some aspect of the correlation matrix (the ones unimportant to my substantive question) are not well modeled? Am I still safe in saying that one causal structure is more likely than the other?</p>
http://openmx.psyc.virginia.edu/thread/2357#commentsGeneral SEM DiscussionsMon, 30 Sep 2013 19:45:13 +0000fife2357 at http://openmx.psyc.virginia.edulatent variable
http://openmx.psyc.virginia.edu/thread/2356
<p>HI<br />
I want to estimate a sem models with binary data<br />
all exemple that i see it i find covariance between the two latent variable </p>
<p>but in sem the arrow have one head i dont want covariance between latent variable it is a regression i think </p>
<p>i use matrix method<br />
how can i change cov between latents variable in matrix </p>
<p>please help me<br />
thanks<br />
emmy </p>
http://openmx.psyc.virginia.edu/thread/2356#commentsGeneral SEM DiscussionsMon, 30 Sep 2013 14:12:02 +0000emmy2356 at http://openmx.psyc.virginia.eduBINARY DATA
http://openmx.psyc.virginia.edu/thread/2332
<p>Hello, everyone, i am a novice at SEM,<br />
i want to test a hypothesis with structural equation model<br />
all my observed data are binary<br />
it is any pretreatment should i do<br />
which method of estimation should I apply<br />
please please help me</p>
http://openmx.psyc.virginia.edu/thread/2332#commentsGeneral SEM DiscussionsFri, 20 Sep 2013 11:31:28 +0000emmy2332 at http://openmx.psyc.virginia.eduAMOS Design Question
http://openmx.psyc.virginia.edu/thread/2289
<p>Hello everybody!</p>
<p>I am new here and I would like to start of by asking you a question regarding a design I made in AMOS. It is meant as part of my graduation thesis in Supply Chain Management.</p>
<p>I am following the structural model of one paper, which uses first- and second-order factors, however, it reports only a simplistic version of this model and it is hard for me to identify if I replicated the model well. </p>
<p>Information about my model:<br />
- My independent latent variables are IC, CO and IG;<br />
- My dependent latent variables are C, Q, FLEX, INN and FIN;<br />
- My mediating variable and at the same time a second-order factor is SCC measured by latent variables: IS, T, IA and JD, </p>
<p>Information about article's model:<br />
- Article's independent latent variables are: Interaction, Coordination, Integration<br />
- Articles dependent latent variables are: Cost, Quality, Flexibility, Innovativeness<br />
- The article's mediating and 2d-order factor is Buyer-supplier collaboration and according to the article it is measured by latent variables information sharing, joint decision-making and incentive alignment. </p>
<p>I would like to know if the input of my model into Amos was correctly done as the article does not show how measures of the second order factor were built into the model. In the attachment you may see the article's simplistic model and the version of my model built in AMOS. </p>
<p>Furthermore, the article reports the following test, which I did not come across in your videos. Quote:</p>
<p>"Statistical evaluation of the second-order factor model can be evaluated by comparing the fit of the first-order model with that of the more restrictive second-order model. The target coefficient (T), which is the ratio of the x2 for the first-order model to the x2 of the second-order model, provides an indication of the degree to which the second-order model accounts for the relations among the first-order factors (Marsh and Hocevar, 1985; Tanriverdi and Venkatraman, 2005). Conducting confirmatory factor analysis and evaluating both a first-order and second-order factor model yields a (T) coefficient of 0.984."</p>
<p>I would like to know whether this paper ran the first-order model without including Buyer-supplier collaboration as a second-order factor, meaning that the latent variables information sharing, joint decision-making and incentive alignment were direct mediators between the independent and dependent variables? And afterwards with Buyer-supplier collaboration and used the ratio of two chi-squares?</p>
<p>Any ideas are highly welcome. Many thanks in advance.</p>
<p>My sincere thanks,</p>
<p>Hans</p>
<table id="attachments" class="sticky-enabled">
<thead><tr><th>Attachment</th><th>Size</th> </tr></thead>
<tbody>
<tr class="odd"><td><a href="http://openmx.psyc.virginia.edu/sites/default/files/amos model.png">amos model.png</a></td><td>175.56 KB</td> </tr>
<tr class="even"><td><a href="http://openmx.psyc.virginia.edu/sites/default/files/Screen Shot 2013-08-22 at 9.42.51 PM.png">Screen Shot 2013-08-22 at 9.42.51 PM.png</a></td><td>112.16 KB</td> </tr>
</tbody>
</table>
http://openmx.psyc.virginia.edu/thread/2289#commentsGeneral SEM DiscussionsThu, 22 Aug 2013 21:10:35 +0000Hansjansen2289 at http://openmx.psyc.virginia.eduSuppression: Interpretation
http://openmx.psyc.virginia.edu/thread/2269
<p>I'm using AMOS. I have 3 predictors and one dependent. they are all positively correlated with each other. However one predictor has a negative regression weight, which is somewhat illogical. Removing that predictor causes the other IV regression weights to drop, so it seems that it is having a suppression effect.</p>
<p>The means of my dependent variable are pretty high. So if it is having a negative effect, it isn't very strong, despite the fairly high negative regression weight (-0.6). The positive regression weights of the other two IVs are 0.5 and 0.9 respectively. I'm trying to figure out how to interpret this result. Is it possible that the positive IVs are simply overpowering the negative IV? I'm an SEM newbie.</p>
<p>Just as an experiment, I removed the covariance lines between my IVs. The two positive IVs saw a light regression weight drop, but the negative regression weight dropped to -0.13. That seems significant but I have no idea what it means or even if that is a valid thing to do. I know I can't leave them like that but I just wanted to see what would happen.</p>
<p>Any advice is appreciated.</p>
http://openmx.psyc.virginia.edu/thread/2269#commentsGeneral SEM DiscussionsSun, 11 Aug 2013 12:59:16 +0000billmac2269 at http://openmx.psyc.virginia.edu