LCA with distal outcome? PreviousNext
Mplus Discussion > Latent Variable Mixture Modeling >
 Jennifer Moren Cross posted on Friday, May 25, 2007 - 8:43 pm
1) Is it possible to estimate LCA with a distal outcome in a single step in Mplus(along the same lines one does with LCGA or GGMM)? (Are there examples of this anywhere?)

2) Would I obtain the same results if I used a two-step procedure whereby, first, I estimated conditional latent classes with covariates (while also identifying my distal outcomes using the "auxillary" command), export this to another statistical package like Stata, and second, run regular regressions of my distal outcomes on class probabilities in Stata?

Much thanks for any information!
 Linda K. Muthen posted on Saturday, May 26, 2007 - 6:31 am
1. Yes. It would be like Example 8.6 but with an LCA model instead of a GMM model.

2. It is always preferable to do the entire model estimation in one step. Doing it in two steps would introduce estimation errors and the standard errors will be incorrect.
 Jennifer Moren Cross posted on Saturday, May 26, 2007 - 11:04 am
Thank you, Linda. Are there any references you would recommend for me to read more about this type of estimation error? Thanks!
 Linda K. Muthen posted on Monday, June 04, 2007 - 11:10 am
I don't know of any specific reference. Perhaps you could find something in a general statistical text.
 Michelle Maier posted on Thursday, March 03, 2011 - 8:57 am
I am running a LCA with several covariates and distal outcomes. I would like to control for the distal outcome scores in the fall so I have specified "outcome on fallscore" in my overall model. Because of this, I get intercepts for my distal outcomes instead of means. I have two questions:

1) What is the best way to request the class means for the outcomes? Tech7?

2) What is the best way to compare my 3 classes on the distal outcomes? I have used MODEL TEST but isn't that comparing the class intercepts and not the means in my particular model?

Thank you in advance for your time!
 Linda K. Muthen posted on Friday, March 04, 2011 - 11:06 am
1. TECH7 is sample statistics. TECH4 and the RESIDUAL option of the OUTPUT command give model estimated values.

2. You would need to define the means in terms of the model parameters using new parameters in MODEL CONSTRAINT where you can also test the differences.
 Keri Jowers posted on Tuesday, September 20, 2011 - 12:08 pm

I'm aiming to use LCA to predict both categorical and continuous distal outcomes. From what I see in Example 8.6, it seems like the current recommendation is to basically do the equivalent of including it as a covariate (c on x). Is that the correct interpretation?

If so, when I do that, the size of my 3 class shift more than I'd expect or like for them to (even when I've specified the stat values for each of the latent class indicators. My understanding is that this indicates that the model may be unstable or not replicable, and that the number of classes may not be correct. The 2- and 3-class models have very similar fit indices:

2-class (79.14%, 20.86%):
AIC = 4431.098
BIC = 4498.885
SSABIC = 4451.258
LMR-adj LRT p = .0007
BLRT p = .0000
entropy = .789

3-class (47.31%, 31.33%, 21.36%):
AIC = 4400.205
BIC = 4504.146
SSABIC = 4431.146
LMR-adj LRT p = .0378
BLRT p = .0000
entropy = .720

The 3-class model is a better fit theoretically/substantively. What, then, is the best way to estimate the association between the latent classes and the distal outcome? Can I trust the results I get when the classes are shifting?

Thanks so much in advance for your input!
 Bengt O. Muthen posted on Tuesday, September 20, 2011 - 6:51 pm
In ex 8.6 the distal outcome is u, not x.
 Keri Jowers posted on Wednesday, September 21, 2011 - 10:43 am
Apologies -- I mistyped. So, in ex 8.6, the distal is mentioned only in the "categorical = u" statement. How would one estimate the association between the latent classes and a continuous distal?
 Linda K. Muthen posted on Wednesday, September 21, 2011 - 11:14 am
The key is that u is on the NAMES list so it is an analysis variable. In the case where all variables on the NAMES list are not analysis variables, u would have been on the USEVARIABLES list. The same holds for a continuous distal outcome. It needs only to be on NAMES or NAMES and USEVARIABLES.
 Keri Jowers posted on Thursday, September 22, 2011 - 11:49 am
Right, I've got the NAMES and USEVARIABLES piece. Perhaps a better way of posing my question is this: When the continuous distal is in the USEVARIABLES list and I then set the start values for my LC indicators (not the u) using my previously obtained thresholds to try to preserve my classes, the output provides me with class-specific means for the distal, and these are based on the re-estimated model I mentioned above. Are these means and their associated p-values intended to be interpreted as the association between the latent class and the distal? This seems counterintuitive to me.
 Linda K. Muthen posted on Thursday, September 22, 2011 - 1:16 pm
The relationship between the categorical latent variable and the distal is found in the varying of the means of the distal across classes. The question you want to ask is if these means are the same across classes or different. You can use MODEL TEST to answer this question.
 Keri Jowers posted on Friday, September 23, 2011 - 8:07 am
Thanks so much! One final question -- how concerned should I be that the class sizes change drastically when compared to when the distal is not included in the USEVAR statement? Not only are the class proportions very different (below), but the sample proportions within each class are very different:

without distal in USEVAR:

with distal in USEVAR:
 Linda K. Muthen posted on Friday, September 23, 2011 - 9:45 am
When you add a distal outcome, it is the same as adding another latent class indicator. The classes will be affected by this. This means that the distal is being taken into account which it should be.
 Debbie Hahs-Vaughn posted on Monday, February 27, 2012 - 7:58 pm
We have estimated a six-class solution in LPA and are interested in using the LPA class membership (which was estimated at Time 1) to predict a distal outcome (at Time 2) while controlling for various attributes at Time 1. I recently attended a conference and heard a presentation that discussed "distal-as-consequence" where class membership is treated as missing data in the regression of the distal outcome on class membership and multiple imputations are used based on the posterior class probabilities (obtained from the estimated growth mixture model without the distal outcome included) to estimate the association between class membership and distal outcomes.

I have searched the Mplus archive and reviewed papers posted as well as the manual for more information/examples on this. Thus far, I have not found any. Any suggestions? Thank you.
 Bengt O. Muthen posted on Monday, February 27, 2012 - 8:14 pm
Section 4 of this paper on our web site discusses plausible values for latent class variables obtained by multiple imputation and how those plausible values can be used:

Asparouhov, T. & Muthén, B. (2010). Plausible values for latent variables using Mplus. Technical Report.
 C. Gantz posted on Monday, February 02, 2015 - 7:48 pm
I have read with great interest the many posts on using LCA to predict distal outcomes. I understand this is a complex topic.

In my analysis, I would like to use a 3 class solution at T1 to predict a variety of T2 continuous outcomes. I additionally would like to control for T2 outcomes at T1. In the first step of this analysis, a three class solution was best based on AIC, BIC and Lo-Mendell, with entropy of .89. These three classes also make a lot of sense theoretically.

My question is as follows:

I understand that often the one step approach is preferable here. However, I read the Clark & Muthen (2009) piece, and it seems that when entropy is high, it is acceptable to use the most likely class membership. When I included the outcome in the class estimation, this significantly changed the formation of the latent classes in a way that no longer made theoretical sense. Am I right to interpret the Clark & Muthen (2009) paper that in this case, given the high entropy from my 3 class results, I could be justified to assign most likely class membership and use these in follow up analyses?
 Bengt O. Muthen posted on Tuesday, February 03, 2015 - 10:26 am
 C. Gantz posted on Tuesday, February 03, 2015 - 10:32 am
Thank you so much for your quick reply, Bengt!
 C. Gantz posted on Tuesday, February 03, 2015 - 10:40 am
Apologies, one quick follow up here: The Clark & Muthen paper appears to be unpublished - is this the case? If so, do you happened to have a reference to this line of logic in a published paper?

Thank you!
 Linda K. Muthen posted on Wednesday, February 04, 2015 - 9:54 am
You can open the data file with any Mac text editor.
 Bengt O. Muthen posted on Wednesday, February 04, 2015 - 11:02 am
I think you can refer to the paper below for this logic:

Asparouhov, T. & Muthén, B. (2014). Auxiliary variables in mixture modeling: Three-step approaches using Mplus. Structural Equation Modeling: A Multidisciplinary Journal, 21:3, 329-341. The posted version corrects several typos in the published version. An earlier version of this paper was posted as web note 15. Appendices with Mplus scripts are available here.
 Yueqi Yan posted on Monday, April 27, 2015 - 9:42 pm
Is there any way to examine effect size when comparing the mean differences of the distal outcomes among classes?

 Bengt O. Muthen posted on Tuesday, April 28, 2015 - 10:23 am
You can divide the mean difference by the standard deviation of the distal.
 Yueqi Yan posted on Wednesday, April 29, 2015 - 12:10 pm
Thanks Bengt! So how I can receive the standard deviation of the distal. My distal is latent variable. I used fixed factor loading method and could not directly receive the variance information for the latent distal from the output. There is only standard error coming up with mean difference and residual variance of the distal outcome for each latent class. Should I run a separate model without latent class to see the variance of the distal outcome? Thanks again!
 Bengt O. Muthen posted on Wednesday, April 29, 2015 - 6:00 pm
You get the factor variance of a distal in TECH4.
 Carol Rhonda Burns posted on Friday, October 23, 2015 - 6:23 am

I am running a LCA with regression using the 3 step, where the Latent Classes predict distal outcomes.

Is it possible to control for other variables in the model that are not used to form the Latent classes such as ethnicity or ses?

many thanks
 Bengt O. Muthen posted on Friday, October 23, 2015 - 3:12 pm
Yes, manually - see Web Note 21.
 Carol Rhonda Burns posted on Friday, October 23, 2015 - 4:02 pm
Thank you
 Sabrina Thornton posted on Saturday, October 24, 2015 - 1:31 pm
I am trying to use a 3 class latent profile model to predict a continuous distal variable. I was only able to proceed to LPA with an auxiliary variable included in the assessment, following the four articles I have read:
1. Clark & Muthén (2009). Relating Latent Class Analysis Results to Variables not Included in the Analysis
2. Asparouhov & Muthén (2014). Auxiliary variables in mixture modeling
3.Lanza et al. (2013). Latent Class Analysis With Distal Outcomes
4. Web note 14
I decided to go for the Lanza approach, following the simulation results from article 2, in which it states that Lanza approach seems to perform well particularly in the situation of my proposed model. Here is the syntax for the LPA using Lanza:

classes = c(3);
auxiliary = PRODIS(dcon);
type = mixture;
savedata: file is LCAinput1.dat;
save = cprob;

I know that I need to use the data saved from the above to run the next analysis, using 3 class to predict the continuous distal variable. I tried to imitate ex8.6, but I do not understand how I can use it in the context of my model. Your help will be very much appreciated.
 Bengt O. Muthen posted on Saturday, October 24, 2015 - 5:04 pm
You can choose between 3 ways to handle the distal: 1-step, automatic 3-step, and manual 3-step. Your input is for automatic 3-step. This does not need a further step, but you get all the information you need in the output - the distal's means and tests of their differences across classes.

In the 1-step approach the distal is part of the model, that is, on the USEV list. This gives you the means of the distal in different classes.

You should add a 5th paper to your reading list, which shows the methods we recommend in 2 tables at the end:

Asparouhov, T. & Muthén, B. (2014). Auxiliary variables in mixture modeling: Using the BCH method in Mplus to estimate a distal outcome model and an arbitrary second model. Web note 21.

For a continuous distal we now recommend "BCH".
 Sabrina Thornton posted on Sunday, October 25, 2015 - 5:13 am
Hi Bengt,

This is massively helpful. I have read the web note 21, and attempted to run the analysis using automatic BCH approach. I presume that this is a new feature developed after version 7.2 since I got an error message saying:

*** ERROR in VARIABLE command
Unrecognized setting in the AUXILIARY option: BCH
for variable(s): PRODIS

 Linda K. Muthen posted on Sunday, October 25, 2015 - 6:41 am
BCH came out in Version 7.3.
 Michelle Mack posted on Wednesday, January 27, 2016 - 5:31 pm

I created profiles of families based on their responses to items concerning their relationship. I am now interested in using these profiles to predict distal outcomes for the siblings in those families (data is nested).
Does the manual BCH method work with multi level modeling with continuous outcomes? I was told that perhaps I need to use BCH weights that are generated in the latent class run but I'm not really sure what this means. If this is correct can you please clarify.

 Bengt O. Muthen posted on Friday, January 29, 2016 - 6:23 pm
BCH is suitable for distal outcomes, but is not available for multilevel situations.

But having siblings in families does not necessitate multilevel modeling - instead you can use a "wide" approach. See our Topic 3 or 4 handouts.
 Massimiliano Orri posted on Friday, February 12, 2016 - 2:26 am
Dear Drs Muthén,

I performed a LPA having 2 latent variables (3 profiles each), and I would like to predict a distal continuous outcome using the BCH approach (manual, I’d like to have other covariables in my final model).

- It is possible? I tried, and by imput is correctly read, calculation are done, but the output is empty (there is nothing after the "warnings" section). Also, BCH weights are not computed (although the file has been created, it's empty).

- If BCH is not available in my case, do you suggest an alternative approach (other than categorize subjects using latent class membership)?

Thanks for your help
 Bengt O. Muthen posted on Friday, February 12, 2016 - 6:36 am
Please send your input, output, and data to Support along with your license number.
 Myungho Shin posted on Sunday, April 10, 2016 - 7:26 am
Dear Muthen,

I tried to compare estimated within-class means of the distal outcome by DE3STEP, DU3STEP, DCON, and BCH.

Below is the part of input file I used in the practice simulation study (from Appendix P, Asparouhov, T., & Muthen, B. (2014)):

Names are u1-u5 y;
Generate = u1-u5(1);
Categorical = u1-u5;
Genclasses = c(2);
Classes = c1(2);
Nobservations = 500;
Nrep = 10;
Auxiliary = y(DE3STEP);

I have found differences estimated within-class means of distal outcome from the results "EQUALITY TESTS OF MEANS ACROSS CLASSES USING THE (3-STEP / BCH / ...) PROCEDURE".

1) I wonder whether the generated data in each simulation conditions (3STEP, DCON, BCH) are the same?

2) If those generated data are not the same, is it possible to say the results can be compared across the approaches?

Thanks in advance for your help.

Myungho Shin
 Tihomir Asparouhov posted on Monday, April 11, 2016 - 1:29 pm
1) you can save the generated data using
REPSAVE = ALL; SAVE = sample*.dat;

2) if the data is not the same most likely you should not compare the results.
 Myungho Shin posted on Monday, April 11, 2016 - 5:52 pm
Thanks for your help.

I have additional questions on my practice simulation.

1) I performed 4 separate monte carlo simulation. changing the approaches with other conditions remaining equal as below;

study 1

"auxiliary = y(DE3STEP);"

study 2

"auxiliary = y(DU3STEP);"

study 3

"auxiliary = y(DCON);"

study 4

"auxiliary = y(BCH);"

Within each study, I got different data sets generated. Is it still possible to compare performances of the approaches?

2) I tried to conduct external monte carlo simulation in order to have the same data set across the approaches simulated. However, I got an error messeage:

*** ERROR in VARIABLE command
Auxiliary variables with E, R, R3STEP, DU3STEP, DE3STEP, DCATEGORICAL, DCONTINUOUS, or BCH are not available for TYPE=MONTECARLO.

It would be appreciated if you could give ant advice on this.
 Tihomir Asparouhov posted on Tuesday, April 12, 2016 - 10:29 am
1) I get the same data sets. Send your example to

2) You can use Mplus with R to do external montecarlo

or this
 irene Spencer posted on Thursday, April 21, 2016 - 10:54 am

If I restructure my data that has nesting within person and within family (i.e., longitudinal data collected from 2 siblings in a family)to the wide format in order to use the BCH method for predicting distal outcomes, would I would need to estimate the distal outcome for each sibling separately (i.e., class membership predicting an outcome for first born siblings and then for second born siblings?)

 Bengt O. Muthen posted on Saturday, April 23, 2016 - 9:33 am
If the distal outcome is a sibling-level variable, yes. But not if the distal outcome is a family-level variable.
 Kaisa Perko posted on Thursday, October 06, 2016 - 2:47 am

I have tried to figure out what would be the best method for analyzing differences in distal outcome variables in my case. It seems to me that the Lanza method (DCON) would work best in my data (with BCH I encounter problems). However, I am aware of the limitations of my statistical understanding that my questions also reflect. Expert advice would be greatly appreciated to be able to gain results that can be trusted.

1) Concerning how the Lanza method is implemented in Mplus,
1A) is it ok to analyze multiple distal outcome variables? I became perplexed when I read from Bakk & Vermunt (2016) that "A limitation of the LTB [Lanza, Tan, & Bray, 2013] approach is that it cannot be used with multiple distal outcomes. A possible way out is to repeat the LTB analysis for every distal outcome, but in doing so there is no guarantee that the LC solution will remain the same across analyses."
1B) How can I know if the class solution changes across the analyses for multiple distal outcome variables (with DCON specifically)?

2) I wonder if there is a way to test whether within-class changes (e.g., from T1 to T2) in the distal outcomes (auxiliary variables) are statistically significant? I mean comparable to the time effect in repeated measures ANOVA. Maybe with the Wald test within classes but how to implement the test to auxiliary variables?
 Bengt O. Muthen posted on Thursday, October 06, 2016 - 10:47 am
I would try to get BCH to work. If you like, send to Support along with license number.

2) You can do that using either Model Constraint to express differences, or with Model Test. Don't know what you mean by distal outcomes (aux vbles).
 Kaisa Perko posted on Friday, October 07, 2016 - 2:48 am
Thank you very much for the response (Oct 6th 10:47), which prompted me to reconsider BCH. After leaving out a problematic variable also the problems with BCH disappeared. So, I can use BCH instead of DCON.

I have understood that with BCH, the distal outcome variables should follow normal distribution. (With distal outcomes I mean variables that are compared across the latent classes and that are not part of the mixture model -I am sorry if I am confusing the terminology).

1) When using BCH, does Mplus inform me if the distribution deviates too much from normality?
2) With non-normally distributed variables, should I use DCON instead?
 Bengt O. Muthen posted on Friday, October 07, 2016 - 9:47 am
I don't think the distal distribution needs to be exactly normal. Stay with BCH.
 Jordan davis  posted on Friday, October 14, 2016 - 12:56 pm
I am attempting to link an LCA with a growth mixture model (Nylund-Gibson et al., 2014). I have run the three step method for the LCA and the GMM and looked at distal outcomes using the AUXILIARY = XXXX(BCH) command. However, when I link the two models using

C2 on C1;

I get an error when I use the AUX command to look at distal outcomes.
*** ERROR in VARIABLE command
are not available with TYPE=MIXTURE with more than one categorical latent variable.

I attempted to simply regress my distal outcome on my categorical latent variables C1 and C2 - and the output provided results by transition pattern. I'm assuming the Intercepts here are the means when transitioning from class X to class X - however, I'm unsure if this is the appropriate way to look at a distal outcome when linking these two models.

 Bengt O. Muthen posted on Friday, October 14, 2016 - 2:26 pm
You should take a look at the 3-step LTA approach which is similar to your case of two latent class variables. See the paper on our website and its Appendices L, M, N:

Asparouhov, T. & Muthén, B. (2014). Auxiliary variables in mixture modeling: Three-step approaches using Mplus. Structural Equation Modeling: A Multidisciplinary Journal, 21:3, 329-341. The posted version corrects several typos in the published version. An earlier version of this paper was posted as web note 15. Download appendices with Mplus scripts.
 mac posted on Tuesday, February 21, 2017 - 12:58 pm

I am using a manual 3-step approach to predict a distal outcome controlling for demographic covariates. My specific question is whether my code is assuming unequal (a la DU3STEP) or equal variance (DE3STEP) across class-specific distal outcome means. Your assistance is much appreciated.

File is STEP3DATA.dat ;
Listwise is on;
Names are



Classes = class(3);
Idvariable is PID;

Missing is * ;

STARTS = 1000 100;


class on X1 X2 X3;

[ModalC#1@1.735 ModalC#2@-0.394];
[DISTAL] (M1);

[ModalC#1@-1.492 ModalC#2@2.336];
[DISTAL] (M2);

[ModalC#1@-3.729 ModalC#2@-3.119];
[DISTAL] (M3);

Model test:
 Bengt O. Muthen posted on Wednesday, February 22, 2017 - 11:59 am
You see this in the output - but judging from your input the variances are not specified to be different across classes; so DE3STEP.

See also Table 7 in our web note 21.
 mac posted on Wednesday, February 22, 2017 - 12:50 pm
Thank you so much for your help, Dr. Muthén. I really appreciate it. I have one follow-up question if I may. How would you modify the code to allow unequal variances (DU3STEP)? I gather there is a change to the corresponding "[DISTAL] (M);" line for classes 1, 2, and 3. Thank you again.
 Bengt O. Muthen posted on Wednesday, February 22, 2017 - 6:05 pm
Just mention the variance in each class. Like:

 Jamie Taxer posted on Wednesday, June 14, 2017 - 4:03 pm
I am using the BCH approach to examine differences in distal outcomes across latent profiles. A reviewer asked me to report the standard deviation of each distal outcome for each latent profile in addition to the mean and standard error. Is there a way for me to calculate the standard deviation of each outcome for each group? I already tried Tech4 and the output is blank.
 DavidBoyda posted on Sunday, December 17, 2017 - 10:50 am
Dear Dr Muthern,

I have am modeling an anuxillary model with a distal outcome. My distal outcome is categorical but MPLUS is reporting the variable as continuous, so I declared the Y in the usevar command and in the model command:

but i recieve errors.

Advice would be appreciated.

c on b1 ;
y on C ;

[n#1@13.801 n#2@9.567];
[y ] ;

[n#1@-0.349 n#2@0.459] ;
[y] ;

[n#1@-2.437 n#2@-4.363] ;
[y] ;
 Bengt O. Muthen posted on Sunday, December 17, 2017 - 3:47 pm
You don't say "y on c" - this is captured instead by the mean of y changing over the c classes. Also, if you declare y at categorical you should not refer to means (that is [y]) but thresholds (that is [y$1]).

If this doesn't help, send your output to Support along with your license number.
 Jung Yun Jang posted on Tuesday, May 22, 2018 - 1:18 pm
Dear Drs. Muthen,

I have run a Latent Profile Analysis on my dataset and found a 4-class solution to be best fitting. Next, I would like to run a survival analysis across these 4 classes to see how they might differ in their time to an event. Considering this time variable as a distal outcome, can I use the BCH method to ensure that the outcome does not affect the findings of the LPA? If yes, what would I do in terms of setting up a model, using the BCH method?

I hope this question makes sense and look forward to your guidance.

Thank you!
 Tihomir Asparouhov posted on Wednesday, May 23, 2018 - 1:27 pm
Take a look at Section 3 in
and User's Guide example 8.17
 Tihomir Asparouhov posted on Wednesday, May 23, 2018 - 1:35 pm
Also read page 684 from the manual.
 EUN JUNG LEE posted on Tuesday, August 21, 2018 - 12:53 am
Dear. Dr. Muthen.

I have conducted my study using LCGA. I got 3 classes through unconditional modeling. Now I am trying to examine distal outcomes on classes longitudinally. So I would like to use LGM to do it by fixing the measurement parameters of the latent class at values from the unconditional model(utilizing 3 step process by Vermunt, 2010)

I am not sure whether my syntax is right for the purpose. Please review my syntax here:
(d2012-d2015 are outcome variable, n=latent classes)

VARIABLE: names = k2012 k2013 k2014 k2015 f2012 f2013 f2014 f2015 d2012 d2013 d2014 d2015 cpr1 cpr2 cpr3 n;
USEV = n d2012-d2015;
classes = c(3);
MISSING = all(999);
analysis: type=mixture;
model : %overall%
i s | d2012@0 d2013@1 d2014@2 d2015@3;

Thank you in advance.
EJ Lee.
 Bengt O. Muthen posted on Tuesday, August 21, 2018 - 6:03 pm
I don't see your distal variable in this setup. Use the BCH method described in our Web Note 21 on our website.
 EUN JUNG LEE posted on Wednesday, August 22, 2018 - 3:58 am
Thank you for your advice, Dr. Muthen.

As I mentioned my previous post, I'd like to get trajectories of outcome variable according to each class, after getting 3 class using LCGA.
I am a beginner at MPlus, so I don't know exactly how to set outcome variable up longitudinally into the syntax.

After reading web note 21, I made new syntax.
I am still not sure if it is right one.

The syntax is here:

title: outcome examination 2step
data: file = BCH2.dat;
VARIABLE: names= k2012 k2013 k2014 k2015 f2012 f2013 f2014 f2015 d2012 d2013 d2014 d2015 W1 W2 W3;
USEV = D2012-D2015 W1-W3;

model: %overall%
I S|D2012@0 D2013@1 D2014@2 D2015@3;
model test:
output:sampstat stdyx;

Your advice is urgently needed.
Thank you so much.

 Bengt O. Muthen posted on Wednesday, August 22, 2018 - 4:48 pm
I still don't see your distal outcome - look at the "manual" approach in Section 3.2 of Web Note 21.
 EUN JUNG LEE posted on Wednesday, August 22, 2018 - 9:32 pm
Followed my prior post,

Thank you so much for your detailed advice.
I am sorry to bother you.
I still have trouble with my syntax.

Actually, my syntax was renewed one based on 'manual approach' in section 3.2 of web note 21. It was a syntax for the second step.
I got 3 classes and BCH weight implementing 'manual 1st step'.

My distal outcome variables are d2012-d2015. The part of "model" command was for distal outcomes.

model: %overall%
I S|D2012@0 D2013@1 D2014@2 D2015@3;

Seeking for trajectory of each class, I set up a command for LGM of d2012-2015.

Using this syntax, I got some results similar to ones from a model which has fixed classes without 3-step approach.

I am wondering this way is right.
I look forward to your guidance.

Thank you so much.
 Bengt O. Muthen posted on Friday, August 24, 2018 - 5:36 pm
If your distal outcome variables are d2012-d2015, what are the outcomes that determine your classes in Step 1?
 EUN JUNG LEE posted on Friday, August 24, 2018 - 7:06 pm
Thank you, Dr. Muthen.

Here' the step 1 syntax.

VARIABLE: names= id k2012 k2013 k2014 k2015 f2012 f2013 f2014 f2015 d2012 d2013 d2014 d2015;
USEV = k2012-k2015 f2012-f2015
AUXILIARY = D2012-D2015;
starts=500 10;
model: %overall%
i_k s_k | k2012@0 k2013@1 k2014@2 k2015@3;
i_f s_f | f2012@0 f2013@1 f2014@2 f2015@3;

k2012 with f2012;
k2013 with f2013;
k2014 with f2014;
k2015 with f2015;

I tried multidimensional LCGA at the step 1 using k2012-k2015 and f2012-2015 to determine classes.

I really appreciate for your guidance.
 Bengt O. Muthen posted on Saturday, August 25, 2018 - 3:00 pm
The input you show in your Aug 22, 3:58am post looks correct based on the information you have now given me. Note that the mean of both i and s will be different for different classes.
 EUN JUNG LEE posted on Sunday, August 26, 2018 - 12:56 am
Dear, Dr. Muthen.

Thanks to you, I was able to go a step further.
I greatly appreciate for your help.

 Silvia Colonna posted on Monday, December 17, 2018 - 2:09 pm
Dear all,

I am trying to run an LPA with a predictor and a distal outcome and I have a couple of questions just to check if I have understood this properly.

1) I can't use an automatic 3 step approach for my predictor (R3STEP) and my distal outcome (BCH) in the same model. If I want to do that, I need to run two different models. As long as the classes estimated do not change, I can interprete my results as if it was one model including both a predictor and distal outcome. Am I correct?

2) I am actually interested to see if my latent profiles predict my distal outcome using ON statements. To do that, I need to use a manual approach constraining my class parameters as I don't want my original model to change, correct?. I was wondering if there is an example I can look at about constraining the parameters. Also, if I do use these constraints, would I be able to use R3STEP for my predictor in the same model?

Thank you

 Bengt O. Muthen posted on Monday, December 17, 2018 - 4:43 pm
1) - 2) No, you should use the manual approach described in Section 3.2 of web note 21. Web note 21 shows the scripts.
 Silvia Colonna posted on Tuesday, December 18, 2018 - 2:48 pm
Dear Dr Muthen,

thank you for your suggestion, this was really helpful. I just have a follow up question.
The code for the second step of the manual approach works fine when I include my predictor but when I add my distal outcomes I encounter an error that I have tried to overcome, unsuccessfully.

The error says that the ON statements in my Model would be ignored and suggested that this could be due to the fact that I need algorithm integration.

I tried to add the algorithm integration but apparently it is not compatible with BCH. By reading previous posts, I thought that this error could also refer to missing values in my distal outcomes, I tried to remove them but I still get the same error.

Do you have any suggestions?
thank you as usual for your precious help.


ps. Just to specify, as indicated by the webnote, my predictor and distal outcomes are all in the same model. I was trying to run separate ON statement (one by one) to better understand where the error was and what it was referring to.
 Bengt O. Muthen posted on Tuesday, December 18, 2018 - 5:29 pm
Perhaps you are saying Y ON C? That's not allowed - instead, the effect of C on Y is seen in the means of Y being different for the different C classes.

If this doesn't help, send your output to Support along with your license number.
 Silvia Colonna posted on Thursday, December 20, 2018 - 5:03 am
That's exactly what I was trying to do, now it is clearer.
Thank you very much.

Best wishes
 Silvia Colonna posted on Wednesday, January 09, 2019 - 4:04 am
Dear all,

I am still trying to run an LPA with two distal outcomes and a predictor.
I am not sure I have written my code properly, this is my syntax:


NAMES = ID Irr RT_go K L_N ModU Car LL SL LW SW odd cd cdagg;
Auxiliary = cd (BCH) cdagg (BCH);
CLASSES = C (3);

STARTS = 80 16;
type = mixture;

C on Irr;


I am not sure about this syntax as I am having a problem with the classes that are swapped. To clarify, when I was trying to decide on my best model I had, lets say, C1= 4 subjects, C2 = 6 sbj and C3 = 5 sbj; whereas with this syntax I have
C1 = 4 sbj, C2 = 5; C3 = 6 but the people in each class are the same, only the class numbers are swapped. Also the AIC and BIC are slightly different but I thought it could be due to the greater complexity of the model.

Any feedback would be greatly appreciated.

 Bengt O. Muthen posted on Wednesday, January 09, 2019 - 5:01 pm
When you say your "best model", I assume that you mean when analyzing only the latent class indicators.

It is ok to have this class switching.
 Silvia Colonna posted on Thursday, January 10, 2019 - 6:47 am
Yes, that is what I meant!
Thank you for your help.
 DavidBoyda posted on Friday, May 31, 2019 - 10:43 am
Dear Professor,

I have an optimal 2 class model with 2 distal outcomes. Can I check my syntax is correct as my model results have really high P-values (999.000)

Infact, should i be looking at the model results section?

names are
Demands Control ManagerialS PeerS Relationships Role Change
SocialDs Depres Age Sex Ethnic Post CPROB1 CPROB2 N ;

USEVAR are N ;
Nominal = N ;

Classes = c(2) ;
Missing = * ;
Auxiliary = SocialDs (BCH) Depres (BCH) ;

type = mixture ;
start 0;


[n#1@2.479] ;

[n#1@-2.687] ;
 Bengt O. Muthen posted on Sunday, June 02, 2019 - 11:10 am
We need to see your full output - send to Support along with your license number.
 ksk posted on Thursday, November 14, 2019 - 3:16 pm
I'm running a compination of LCA and LPA using binary and continuous indicators. After adding categorical distal outcomes (coded 1, 2, 3), the model didn't run. I wanted to test the assocaition between the class membership and the categorical outcome. Below is my syntax:

USEVARIABLES = y1 y2 y3 y4 y5;
CLASSES = c(4);
NOMIAL = cor6;
LRTSTARTS = 0 0 500 200;

cor6#1 cor6#2 ON c;


The error is below:
The following MODEL statements are ignored:
* Statements in the OVERALL class:
cor6#1 ON C#1
cor6#1 ON C#2
cor6#1 ON C#3
cor6#1 ON C#1
cor6#1 ON C#2
cor6#1 ON C#3

Would you please let me know why the model command didn't work?
 ksk posted on Thursday, November 14, 2019 - 3:27 pm
I also have another question related to the model above.

Instead of the 1-step approach above, can I use the 3-step BCH approach with DCAT? I used the BCH approach when I had continuous distal outcomes predicted by class membership. but I'm not sure when it comes to categorical outcomes.

In addition to the the categorical distal outcome, I want to include a continous covariate to predict class membership.

USEVARIABLES = y1 y2 y3 y4 y5;
CLASSES = c(4);
NOMIAL = cor6;
LRTSTARTS = 0 0 500 200;


LRTSTARTS = 0 0 500 200;

FILE = dcat.csv;

Q2. If the BCH approach works with DCAT, how can I add a continous covariate in the Auxiliary variable list? Thank you!
 Bengt O. Muthen posted on Thursday, November 14, 2019 - 5:14 pm
first post:

You don't have cor6 in your USEV statement.

You should not regress distals on the latent class variable. Instead, those effects are captured by mean/threshold differences in the distal across classes.

second post:

You find answers reading Web Note 21 including its tables 6 and 7.
Back to top
Add Your Message Here
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Options: Enable HTML code in message
Automatically activate URLs in message