st: RE: for Red Owl: assertion is false error: mvencode

[Red Owl <rh.redowl@liu.edu>]

Linda and others who may be interested,

As Nick Cox pointed out, the LCA Plugin version 1.0 (see References below) is not leaving behind the r(gammaSTD) matrix or the r(rhoSTD) matrix requested in your code.  I believe (but am not sure) that these matrices would contain the asymptotic standard errors (ASE) of the latent class membership probabilities (gamma) and ASEs of the item-response probabilities conditional on latent class membership.  

The LCA Stata Plugin user's guide (see References below) indicates in section 3.2 that, "Asymptotic standard errors for LCA parameter estimates are provided when available."  I expect they are just not available for your data, so the plugin is not computing them.

Would you tell us which flavor and version of Stata you are using (e.g., Stata/IC 12.1)?  If you are limited to Stata/IC, you might want to use some R code for which I have developed a model to pre-process the data set in Stata, import it into R and run the poLCA package (see References below), and then output the relevant LCA-related variables into a Stata data set that can be merged into your original Stata data set.  Then you can perform any additional Stata procedures with the LCA-related variables plus your other variables.

To do that, you do not need to be an expert in or to learn R, as the data are pre-processed in Stata and my simple R model code is annotated with clear instructions.  I also have some really basic Stata code to take the csv file output from the R routine and merge the R output variables from poLCA back into the Stata data set.  Of course, R uses a GNU Public License and is readily available for download on most major platforms.

If you would like to pursue that approach, please contact me privately at my university address above.  I don't want to discuss R further on Statalist.  I don't find any prohibition in the Statalist FAQ of mentioning programs other than Stata when they are relevant to a Stata question, but it still seems inconsiderate to me to those who come here to discuss Stata.

If you want to obtain advice directly from the authors of the LCA Stata Plugin, you could email MChelpdesk@psu.edu, as provided on the web site of The Methodology Center at Penn State.

Red Owl


References:

Lanza, S. T., Dziak, J. J., Huang, L., Wagner, A. T., & Collins, L. M. (2011). LCA Stata plugin users' guide (Version 1.0). University Park: The Methodology Center, Penn State. Retrieved from http://methodology.psu.edu

LCA Stata Plugin (Version 1.0) [Software]. (2013). University Park: The Methodology Center, Penn State. Retrieved from http://methodology.psu.edu

Linzer, Drew A. and Jeffrey Lewis. 2011. "poLCA: an R Package for Polytomous Variable Latent Class Analysis." Journal of Statistical Software. 42(10): 1-29. Retrieved from http://www.jstatsoft.org/v42/i10

Linzer, Drew A. and Jeffrey Lewis. 2013. "poLCA: Polytomous Variable Latent Class Analysis." R package version 1.4. Available from http://dlinzer.github.com/poLCA

R-Project at http://www.r-project.org


<previous discussion snipped>
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/