Tech10 standardized residuals
Message/Author
 Jeannie-Marie Leoutsakos posted on Friday, December 29, 2006 - 9:17 pm
Hi,

Sorry if this is a repost, I don't think it went through the first time. In tech10, how are the standardized residuals for bivariate cell counts calculated? It doesn't seem to be (O-E)/sqrt(E)?

thanks,
Jeannie
 Tihomir Asparouhov posted on Tuesday, January 02, 2007 - 11:04 am
The standardized residuals given in tech10 are not the Pearson residuals but are the standardized Pearson residuals. See Agresti's Categorical Data Analysis book, Sections 3.3.1 and 4.5.5. The original article on this topic is The Analysis of Residuals in Cross-Classified Tables, Shelby J. Haberman, Biometrics, Vol. 29, No. 1 (Mar., 1973), pp. 205-220.

The advantage of the standardized Pearson residuals is that they are just like standard normal residuals, while the unstandardized Pearson residuals are less variable. For the bivariate tables these are computed by (O-E)/[sqrt(E)*sqrt(1-E/n)].
 QianLi Xue posted on Sunday, December 07, 2008 - 6:25 pm
Is it true that when sampling weights are used in LCA fitting, the Pearson Chi-square statistics for model fit in the general output and for response pattern frequencies in TECH10 are no longer valid? Is that why some of the Chi-square values in TECH10 become negative?
 Linda K. Muthen posted on Monday, December 08, 2008 - 8:52 am
The response pattern frequencies in TECH10 do not take weights into account. This could be why some of the chi-square values become negative. In the future, TECH10 will take weights into account.
 Lorraine Ivancic posted on Sunday, July 24, 2011 - 9:20 pm
Hi,
I am running a latent class analysis with 10 binary indicators and sample weights. If the tech10 output (the bivariate residuals) does not take account of the sample weights (as mentioned in the post above) how can I assess model fit?
thanks
 Linda K. Muthen posted on Monday, July 25, 2011 - 9:18 am
The current version of Mplus takes the weights into account for TECH10.
 Lorraine Ivancic posted on Monday, July 25, 2011 - 8:28 pm
Thanks Linda. Just to check - by current version do you mean 6.11? I'm currently using version 5.21.
 Linda K. Muthen posted on Tuesday, July 26, 2011 - 10:17 am
Yes, I mean 6.11. I don't know what it was in 5.21. You can run it both with and without weights to see if TECH10 differs.
 LAS posted on Monday, August 01, 2011 - 3:26 pm
Hello. I am running a GMM negative binomial model and I would like to examine the standardized residuals; however, the section of the output where the tech 10 results should appear is blank (except for the heading "technical 10"). Is it possible to obtain standardized residuals with the negative binomial? Thank you.
 Bengt O. Muthen posted on Monday, August 01, 2011 - 5:27 pm
Are you using version 6.11?
 LAS posted on Tuesday, August 02, 2011 - 9:13 am
I was not, but I have since updated my version to 6.11 and it has not resolved the issue. Thank you again.
 Linda K. Muthen posted on Tuesday, August 02, 2011 - 9:28 am
Please send the output and your license number to support@statmodel.com.
 Annette Stelter posted on Tuesday, July 10, 2012 - 12:10 am
Dear Dr. Muthen,
I am running a CFA model using MLR for parameter estimation. The data structure includes a booklet design (i.e., data missing at random), and both 72 categorical indicators (binary responses indicating an ability factor) and 72 continuous indicators (log-transformed response times indicating a slowness factor).
To obtain model fit information for the categorical dependent variables in the model, I wanted to look at TECH10 for the standardized residuals. However, with a higher number of indicators the TECH10 Output disappeared. When I used 6, 9, or 15 indicators per factor TECH10 was regularly provided. With about 30 Items per factor it wasn’t. How do I get TECH10 in this case?
Thank you in advance!
With kind regards,
Annette
 Linda K. Muthen posted on Tuesday, July 10, 2012 - 11:00 am
We don't provide TECH10 for large frequency tables. You would get the following message if this is the case:

TECH10 OUTPUT FOR CATEGORICAL VARIABLES IS NOT AVAILABLE BECAUSE THE FREQUENCY TABLE FOR THE LATENT CLASS INDICATOR MODEL PART IS TOO LARGE.
 Annette Stelter posted on Tuesday, July 17, 2012 - 5:31 am
This message didn´t show up in my outputfile, however just „TECHNICAL 10 OUTPUT“ was written in the last output row:
"TECHNICAL 10 OUTPUT

Beginning Time: 10:25:13
Ending Time: 10:25:30
Elapsed Time: 00:00:17..."

How do I compute the model fit information in my case?
 Linda K. Muthen posted on Tuesday, July 17, 2012 - 7:43 am
Please send the output and your license number to support@statmodel.com.
 samah Zakaria Ahmed posted on Sunday, January 29, 2017 - 11:06 am
what does output of TECH10 of latent class model refer to?
how to interprets?
 Bengt O. Muthen posted on Sunday, January 29, 2017 - 3:16 pm
Study the short course Topic 5 handout and video on our website.
 samah Zakaria Ahmed posted on Monday, January 30, 2017 - 4:34 am
Thanks a lot... I will do
 Bengt O. Muthen posted on Monday, January 30, 2017 - 3:18 pm
ok.
 Daniel Lee posted on Friday, April 28, 2017 - 11:17 am
Hi I'm trying to estimate standardized residuals for a path model. The independent variable is a manifest variable, mediators are latent variables, and the dependent variable is a manifest dichotomous variable.

I was wondering if fit indicators were available when using ML as estimator (logit link) and realized that it was not.

I was told that you can compared observed vs. model estimated frequencies and standardized residuals. When I included tech10 in the output line of the .inp file, I received an error message:

TECH10 OUTPUT FOR CATEGORICAL VARIABLES IS NOT AVAILABLE FOR MODELS WITH COVARIATES.

The error would not go away when I removed the covariates from the model (so all that was left was the one IV, 2 latent variable mediators, and the binary dependent variable). I would appreciate assistance and getting tech10 to work. Thank you!
 Bengt O. Muthen posted on Friday, April 28, 2017 - 3:11 pm
TECH10 would be relevant if you have more than one categorical DV. ML with categorical outcomes does not have fit indices. You can try Bayes which uses probit but the power of it test of model fit has low power. One approach in the ML context is to create neighboring models that are less restrictive than the model you consider and then check if the extra parameters are significant.