Onyx
http://openmx.psyc.virginia.edu/taxonomy/term/43/0
enOnyx, a Structural Equation Modeling (SEM) Graphical User Interface
http://openmx.psyc.virginia.edu/thread/2947
<p>Onyx is a Structural Equation Modeling (SEM) Graphical User Interface.</p>
<p>It is designed to allow users to build and fit SEMs intuitively and fast, to ease the introduction of Structural Equation Models for beginners, and simplify the transition to more complex SEM scripting languages.</p>
<p>Onyx is developed in cooperation of the Mathematical Psychology Lab at UVa and the Formal Methods Project at the Max Planck Institute for Human Development.</p>
<p>Onyx provides import from and export to OpenMx.</p>
<p>Also, publication-quality figures can be exported in a vector representation to LaTeX or in a JPEG image file. Furthermore, Onyx provides a powerful backend for performing maximum likelihood estimation.</p>
<p>Onyx is intended to be a teaching tool for SEM. It facilitates a graphical approach to modeling that also offers an interface to OpenMx and Mplus code. That is, SEMs can be taught in a graphical approach without the need to<br />
teach a specific modeling language. In the progress of a course, a transition from the purely graphical approach to a specific modeling language can be made using exports to OpenMx or to MPlus. An interactive code view allows users to track and compare changes to the model specification in the selected language that are elicited by changes to the graphical representation.</p>
<p>You can download Onyx for free at <a href="http://onyx.brandmaier.de" title="http://onyx.brandmaier.de">http://onyx.brandmaier.de</a>.</p>
<p>You can find a brief user guide at <a href="http://onyx.brandmaier.de/userguide.pdf" title="http://onyx.brandmaier.de/userguide.pdf">http://onyx.brandmaier.de/userguide.pdf</a>. </p>
<p>If you have any questions or issues regarding Onyx, please post here. Or email <a href="mailto:timo@virginia.edu">timo@virginia.edu</a> or <a href="mailto:brandmaier@mpib-berlin.mpg.de">brandmaier@mpib-berlin.mpg.de</a></p>
http://openmx.psyc.virginia.edu/thread/2947#commentsOnyxSat, 26 Apr 2014 22:02:42 +0000tbates2947 at http://openmx.psyc.virginia.eduSuggestion: clones inherit parental model customization
http://openmx.psyc.virginia.edu/thread/4147
<p>Hi,<br />
I am a beginner to both SEM and Onyx..., but see both as highly useful for my research field, plant ecology. </p>
<p>Playing around with a model including many variables, i find that i have to reapply model customization over and again when i compare models with the clone and drag-compare method.</p>
<p>Would it be possible to let clones inherit the customization of the parent ? (path style, show/hide variances etc.)</p>
<p>Many greetings and thanks for a wonderful software !</p>
<p>Thure </p>
http://openmx.psyc.virginia.edu/thread/4147#commentsOnyxTue, 14 Jun 2016 17:19:45 +0000tpha4147 at http://openmx.psyc.virginia.eduDegrees of freedom in Onyx
http://openmx.psyc.virginia.edu/thread/3942
<p>Hi there!</p>
<p>I'm exploring Onyx with great interest as a possible recommendation to my students (instead of AMOS).<br />
In order to familiarize myself with Onxy, I am comparing the results of the Tutorial CFA example with those from AMOS and Mplus. The results in AMOS and Mplus are very similar to each other. Results in Onyx are similar for parameter estimates, but very different for model fit. </p>
<p>It appears that Onyx is calculating the degrees of freedom differently.<br />
In the example, there are three indicators [for which I added one latent factor]. If I remember correctly, that should give me 6 observed (co-)variances: 3*(3+1)/2 = 6. However, Onyx tells me that there are "Observed Statistics: 9.0".<br />
Consequently, the model isn't just-identified (as I expected), but has 3 df.</p>
<p>Why is that the case? Is there a way to make these settings more comparable across software packages?</p>
<p>Any help would be greatly appreciated!<br />
Heiko</p>
http://openmx.psyc.virginia.edu/thread/3942#commentsOnyxTue, 20 Jan 2015 08:43:08 +0000hbreitsohl3942 at http://openmx.psyc.virginia.eduPath analysis
http://openmx.psyc.virginia.edu/thread/3924
<p>Hello,</p>
<p>1. I've built a simple model with Onyx with no latent variables, a constant, one endogenous observed variable and two exogenous observed variables, with one path each from the exogenous variables and constant to the endogenous variable, to see how Onyx may be used for (eventually, simultaneous) regression equations with only observed variables. I created a data set with random values for the two exogenous variables and defined the endogenous variable as the sum of the constant and the sum of the products of coefficient and exogenous variable (Y=const + b1X1 + b2X2). I used 200 observations.</p>
<p>The estimates of all coefficients was accurate, but the fit measures seem off - particularly for such a simple model. Chi-square is 541.05, RMSEA 0.95, CFI 0.70, and TLI 0.41. Only SRMR suggests a good fit (0.02). Are there any reasons for this?</p>
<p>2. In other models with no latents - though with far fewer observations - several fit measures were reported as NaN; TLI turned out negative. What might cause this to happen?</p>
<p>3. Also, at least two text books calculate df as k(k+1)/2 - p where k = # variables and p = # parameters estimated. As I understand it, k(k+1)/2 = number of knowns = number of covariances + the number of correlations. Onyx seems to add k to the formula while reporting observed statistics. What is the reason?</p>
<p>Thanks for your help.</p>
http://openmx.psyc.virginia.edu/thread/3924#commentsOnyxWed, 05 Nov 2014 14:16:20 +0000mdr3924 at http://openmx.psyc.virginia.eduproblem changing path curvatures
http://openmx.psyc.virginia.edu/thread/3895
<p>Hello,</p>
<p>I'm new to using Onyx, and really enjoying it. Right now, my only issue is a cosmetic one.</p>
<p>I am having a lot of trouble changing the paths that I am drawing between variables. Following the manual, I am going into the context menu and selecting "Show Control Points" in order to change this, but when I do so, the control points always appear fixed to the upper left hand corner of the individual model drawing window. It happens this way when working with my main model, as well as any new models I have tried creating; in either situation, the control points are impossible for me to select and drag out of the upper left corner, preventing me from altering the curvature of my paths. </p>
<p>The attached picture shows what I'm talking about. Also, I'm running the Windows version, under Windows 7. Am I missing something, or is this a known issue? Any other thoughts on how to proceed? Thanks!</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/pathProblem.png">pathProblem.png</a></td><td>27.46 KB</td> </tr>
</tbody>
</table>
http://openmx.psyc.virginia.edu/thread/3895#commentsOnyxFri, 05 Sep 2014 14:41:40 +0000Dremalt3895 at http://openmx.psyc.virginia.eduExamples of using Onyx
http://openmx.psyc.virginia.edu/thread/3880
<p>Hello,</p>
<p>I'm a novice in structural equation modeling. I am curious if Onyx works for the following experimental design:</p>
<p>There are 2 treatment variables (T1, with 2 levels; and T2, with 3 levels; each with 3 replicates) and 3 response variables (R1, R2, R3). I'm interested in how each of the T variables affects the R variables, as well as if this relationship changes depending on the level within each T variable. I'm also interested to know how R3 is related to R1 and/or R2.</p>
<p>I would like to repeat this simple design with more replicates and more R variables later. I tried putting the data in Onyx with help from the guide but wasn't sure how to specify some of the variables/paths and other options to see the relationships quantified...</p>
<p>Thanks for any input.<br />
Yev</p>
<p>Here is some sample data with this design:</p>
<p>T1 T2 R1 R2 R3<br />
0 0 7.7 2.8 7<br />
0 0 7.5 1.8 7<br />
0 0 7.3 4.0 8<br />
0 1 7.2 4.9 9<br />
0 1 7.0 22.0 8<br />
0 1 7.0 1.9 7<br />
0 2 7.4 1.3 9<br />
0 2 7.0 1.8 8<br />
0 2 7.4 1.7 9<br />
1 0 5.9 302.6 7<br />
1 0 6.6 141.4 9<br />
1 0 6.7 113.7 7<br />
1 1 7.5 56.2 9<br />
1 1 5.5 89.5 9<br />
1 1 6.0 99.2 8<br />
1 2 7.6 23.9 11<br />
1 2 6.5 74.7 10<br />
1 2 6.6 33.1 12</p>
http://openmx.psyc.virginia.edu/thread/3880#commentsOnyxThu, 21 Aug 2014 16:21:38 +0000ymarusen3880 at http://openmx.psyc.virginia.eduVariance calculations in Onyx
http://openmx.psyc.virginia.edu/thread/2770
<p>I've been trying to match the variances reported by Onyx with those reported by AMOS. Following the example in the AMOS User Guide, I've created a model with four observed variables (recall1, recall2, place1, place2 in the file Attg_yng.xls file) connected to each other with double-headed arrows to determine the covariances of the variables, as is done in AMOS.</p>
<p>The values displayed by Onyx are all different from those displayed by AMOS. For example, the variance of recall1 in AMOS is 5.79 and 110.85 in Onyx. (Excel's VARP function also gives a value of 5.79.) Similarly, the covariance between recall1 and recall2 is 2.56 in AMOS and 121.2 in Onyx (Excel's COVAR gives 2.56).</p>
<p>I'd appreciate an explanation for these differences. Thank you</p>
<p>MDR</p>
http://openmx.psyc.virginia.edu/thread/2770#commentsOnyxTue, 01 Apr 2014 12:37:26 +0000mdr2770 at http://openmx.psyc.virginia.eduOverspecified model
http://openmx.psyc.virginia.edu/thread/2624
<p>Hi,</p>
<p>I'm new to both SEM and Onyx as my question probably shows. </p>
<p>An SEM model that I've specified with Onyx has the symbol with an exclamation mark displayed, and the associated message tells me that the model is overspecified.</p>
<p>Onyx FAQs notes, "If you obtain this error, you have most likely included more freely estimated parameters in your model than there are observed statistics. Try to fix freely estimated parameters or remove paths."</p>
<p>However, Onyx's Estimate Summary reads:<br />
Observed Statistics : 65.0<br />
estimated Parameters : 24<br />
Restricted Degrees of Freedom: 41.0<br />
Why do I get the error warning when the number of freely estimated parameters (24) is less than the number of observed statistics (65)?</p>
<p>MDR</p>
http://openmx.psyc.virginia.edu/thread/2624#commentsOnyxMon, 24 Feb 2014 09:52:42 +0000mdr2624 at http://openmx.psyc.virginia.eduOnyx for producing R scripts
http://openmx.psyc.virginia.edu/thread/2617
<p>Is the an equivalent to Onyx for diagramatically producing R code for data preparation, such as recoding data, creating new variables, matching/merging datasets, etc?<br />
Peter</p>
http://openmx.psyc.virginia.edu/thread/2617#commentsOnyxSat, 15 Feb 2014 23:27:53 +0000cuttance2617 at http://openmx.psyc.virginia.eduOnyx Beta 0.9-534 released
http://openmx.psyc.virginia.edu/thread/1680
<p>Ωnyx (Version 0.9-534), the graphical user interface for SEMs with automatic code generation to OpenMx code is released. You may download Ωnyx for free at onyx.brandmaier.de. Any comments about your experiences with Ωnyx are welcome at this forum. </p>
http://openmx.psyc.virginia.edu/thread/1680#commentsOnyxMon, 22 Oct 2012 18:58:17 +0000timo1680 at http://openmx.psyc.virginia.edu