PPP for Bayesian analysis PreviousNext
Mplus Discussion > Structural Equation Modeling >
 Johnny Zhang posted on Monday, November 14, 2011 - 1:18 pm
I have a question regarding PPP when using the Bayesian estimator. In Muthén, B. & Asparouhov, T. (2011), it says low PPP indicates poor fit but also says a PPP around 0.5 indicates an excellent-fitting model. However, in my analysis, I had a PPP about 0.99. Did this indicate a poor fit model or a good fit model? Thanks.
 Linda K. Muthen posted on Tuesday, November 15, 2011 - 9:31 am
Good fit.
 Rohini Sen posted on Thursday, May 03, 2012 - 7:20 am
If PPP is around 0.36, is that a very poor fit?
 Linda K. Muthen posted on Friday, May 04, 2012 - 10:02 am
This would indicate good fit. You are looking for values greater than .05. A value of .5 would indicate excellent fit.
 Phil Wood posted on Tuesday, November 27, 2012 - 8:35 am
Is it ever acceptable to use the ppp from a Bayesian analysis to compute a Bayes Factor between two models, or is it always preferable to use the BIC? (Assuming you have enough draws to believe it as a point estimator). I read a recent article by Meng in the Annals of Statistics which frowned on doing so, but it seems that argument is based on "using the same data twice," which we also did when using the BIC. Any thoughts from anyone?
 Tihomir Asparouhov posted on Monday, December 03, 2012 - 9:54 am

I haven't seen any methodology on computing BF using PPP. PPP definition typically involves just 1 model while BF involves 2 models. Note also that PPP can be defined in many ways and the way it is defined in Mplus has nothing to do with what is in Meng's article. Mplus uses SEM style chi2 fit function.

If you want to compare PPP and BF testing power and quality I would recommend looking at a simulation study.

 Phil Wood posted on Tuesday, December 04, 2012 - 7:25 am
I had just meant dividing the PPP from one model by the PPP from another model. Looking at just a few calculations, it doesn't sem to work very well in practice relative to using, say, the BIC. Thanks for clearing up my confusion on Meng's article, though!
 Yoonjeong Kang posted on Thursday, April 03, 2014 - 9:30 am

I am running a second-order factor model using both ML estimation and Bayesian estimation. I have 16 subscale scores and 16 subscale scores load on four factors and the four factors load on one higher factor.

I found that Model fits from ML estimation seem to be good based on CFI(0.973) and RMSEA (0.044). However PPP from Bayesian estimation was very low. It was PPP=0.00. I don't understand why two estimation methods provide very different results in terms of model fit. Do you have any idea why this happens?
(For the reference, I used default priors in bayesian estimation.)
 Bengt O. Muthen posted on Thursday, April 03, 2014 - 9:47 am
PPP is more related to chi-square than to CFI. What is your sample size, chi-square value, df, and p-value?
 Yoonjeong Kang posted on Tuesday, April 08, 2014 - 11:21 am
Dear Dr. Muthen,

Sorry for the late response. I have a sample size of 2,200. I got a chi-square value of 689.058, df of 101, and p value of 0.000 using ML.

Thanks a lot in advance for your help.
 Bengt O. Muthen posted on Wednesday, April 09, 2014 - 10:57 am
So the p-value for the chi-square and the PPP agree as is expected. These fit statistics are less forgiving than CFI. So it isn't a matter of Bayes vs ML but a matter of which fit statistics you deem best.
 Yoonjeong Kang posted on Wednesday, April 09, 2014 - 1:59 pm
Dear Dr. Muthen,

Thanks a lot for your answer. I have following-up questions.

I prefer using other fit indices such as CFI and RMSEA than the chi-square when assessing the model fit. It's because the literature has demonstrated that the chi-square is sensitive to sample size.

I wonder whether PPP is also sensitive to sample size or not. If so, I wouldn't want to use PPP to assess model fit. Then my another question is that is there any other model fit indices to assess model fit in Bayesian SEM?

Thanks a lot for your help in advance.

 Bengt O. Muthen posted on Wednesday, April 09, 2014 - 2:13 pm
I don't think one should be so quick to abandon chi-square (or PPP). Instead, try to investigate why the chi-square is not good enough. What's good about the Bayes approach is that you get an interval for the fit and when you relax the model you can see how the lower limit decreases, getting closer to zero (and into the negative).

PPP is based on chi-square and therefore also has its power increase with increasing sample size. Mplus does not give any other Bayes fit measures. Note, again however that I would think you want to relax some restrictions in your model as discussed in

Muthén, B. & Asparouhov, T. (2012). Bayesian SEM: A more flexible representation of substantive theory. Psychological Methods, 17, 313-335.

If you do so, some of your key results might change.
 anonymous Z posted on Tuesday, February 24, 2015 - 1:56 pm
Hi Dr. Muthen,

I just started to use Bayes estimation, and I have two questions.

1. The output showed



Is this a warning/error message?

2. model fit: I got a model fit as below.I assume it is good since p=0.515. But I don't understand the confidence interval info. Is it good or bad?

95% Confidence Interval for the Difference Between the Observed and the Replicated Chi-Square Values

-18.234 17.604

Posterior Predictive P-Value 0.515

Thank you very much!
 Bengt O. Muthen posted on Wednesday, February 25, 2015 - 1:07 pm
1. No, it's a suggestion for good analysis practice.

2. Read my into Bayes paper on our website.
 Jana Holtmann posted on Wednesday, December 02, 2015 - 3:20 am

I've got a question concerning the interpretation of the output on PPP values in Monte Carlo Simulations using Bayesian estimation. The information given on the PPP value is a table with expected and observed proportions, however no percentiles. I understand that the PPP does not follow a known distribution, hence no percentiles? I was wondering how to interpret the expected proportions then. Is it correct if I assume that the expected proportions listed for the PPP are just the possible values of the PPP (and the observed column then gives me the cumulative frequency of values observed in the respective range) ?

 Linda K. Muthen posted on Wednesday, December 02, 2015 - 6:54 am
The description of how to interpret that output is given on page 412 of the current user's guide. This is in Chapter 12 under the title Monte Carol Output.
 Jana Holtmann posted on Wednesday, December 02, 2015 - 7:20 am
Unfortunately the info in the user guide doesn't exactly answer my question. On page 412 it says that "The column labeled Proportions Expected (column 1) should be understood in conjunction with the column labeled Percentiles Expected (column 3)."
But there is no column with Percentiles in case of PPP values.

 Linda K. Muthen posted on Wednesday, December 02, 2015 - 10:15 am
Please send the output and your license number to support@statmodel.com and I will look into this.
 Freya Glendinning posted on Tuesday, June 07, 2016 - 5:24 am
Dear Linda and Bengt,

Please can you help me understand why I get two identical PPp-values and 95% C.I's in two separate models where the IV's are the same but the DV is different?

There are no latent variables in either model and each model has the same number of parameters.

Thanks for your time,
 Bengt O. Muthen posted on Tuesday, June 07, 2016 - 8:53 am
Please send the two outputs to Support along with your license number.
 Freya Glendinning posted on Tuesday, June 07, 2016 - 8:58 am
Hi Bengt,

I am using the universities student license which doesn't cover support.

Is there any more information I can give to help you answer the question on this discussion page?

Thank you,
 Bengt O. Muthen posted on Wednesday, June 08, 2016 - 10:07 am
This can happen if your DV versions are linear transformations of each other. Beyond that, we can only tell from the output and data.
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