Goodness of Fit indicators AIC, BIC, CAIC and EN

vonbergh · Post by **vonbergh** » Tue May 25, 2010 8:21 pm

Hello,

I need to report the GoF AIC, BIC, CAIC and EN of my model. Unfortunately I was not able to find them in SmartPLS. How can I calculate these indicators?

Thank you,
Dennis

vonbergh · Post by **vonbergh** » Sun Sep 11, 2011 6:51 pm

Found out at last:

In the menu: Calculate -> FIMIX-PLS -> apply settings -> Finish

Then:

in the menu: Report -> Default Report (or any other)
in the report: FIMIX-PLS/Quality criteria/Fit Indices

vonbergh · Post by **vonbergh** » Sun Sep 11, 2011 6:54 pm

Good article on GoF:

Wicherts, J.M., Dolan, C.V., (2004) A Cautionary Note on the Use of Information Fit Indexes in Covariance Structure Modeling With Means, Structural Equation Modeling, 11(1), pp45-50.

Available at: http://wicherts.socsci.uva.nl/cautionarynote.pdf

vonbergh · Post by **vonbergh** » Fri Sep 16, 2011 6:05 pm

Check this thread for more info on GoF:

viewtopic.php?t=841&highlight=lisrel

vonbergh · Post by **vonbergh** » Tue Nov 22, 2011 7:44 pm

For a GoF with a threshold of ,50 check page 46 of:

Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(1), 39-50. http://www.jstor.org/stable/3151312

available here

Diogenes · Post by **Diogenes** » Wed Nov 23, 2011 12:19 pm

Hi,

In the p.59 of
ESPOSITO VINZI, V. TRINCHERA, L.; AMATO, S. PLS path modeling: from foundations to recent developments and open issues for model assessment and improvement. In: V. Esposito Vinzi; W. W. Chin; J. Henseler; H. Wang (Eds.); Handbook of partial least squares. p.47-82, 2010. Berlin, Heidelberg: Springer Berlin Heidelberg.

They explain a “relative GoF” to be used in PLS-PM and it was implemented in the XLSTAT-PLSPM (free for 30 day).

Best regards,

Bido

vonbergh · Post by **vonbergh** » Wed Nov 23, 2011 4:52 pm

Hello Bido,

Thank you. XLSTAT-PLSPM looks an interesting program.

I also found interesting info in this thread: viewtopic.php?t=1094

Best regards,
Dennis

MBUFI · Post by **MBUFI** » Thu Jan 19, 2012 3:37 pm

Very useful stuff.

Can the EN from FIMIX be used to assess the PLS goodness of fit?

or does it relate to the segmentation only?

In that case, is the option by Tenenhaus et al. the only way of calculating a GoF?

thanks

mbitar · Post by **mbitar** » Thu Mar 01, 2012 12:47 am

I don't know how to read the following fit indices:

AIC 536.388266
BIC 695.195753
CAIC 695.477444
EN 0.549907

can anyone help please

Taufik · Post by **Taufik** » Sun Apr 08, 2012 2:42 am

mbitar wrote:I don't know how to read the following fit indices:

AIC 536.388266
BIC 695.195753
CAIC 695.477444
EN 0.549907 (EN>0.5 considered unambigous segmentation)

can anyone help please

You need the compare the index values for different segment (K).

Refer to:

Ringle, C.M (2006). Segmentation for Path Models and Unobserved Heterogeneity: The Finite Mixture Partial Least Square Approach. University of Hamrmburg.

Sarstedt, M., Becker, J.-M., Ringle, C. M., & Schwaiger, M. (2011a).
Uncovering and treating unobserved heterogeneity with FIMIXPLS:
which model selection criterion provides an appropriate
number of segments? Schmalenbach Business Review, 63(1), 34–
62.

forum.smartpls.com