Paris, town of love and PLS
A 3-day Course on PLS Path Modelling:
Basic Concepts and Foundations, Advances and Applications
by W.W. Chin, V. Esposito Vinzi, M. Tenenhaus
23-24-25 November 2009
Paris, France
Read more
Today, Michael and I tried out the current beta version of OpenMx. When estimating one of the models we use in our workshops, we found out that Mplus, Amos, Lisrel and OpenMx produce different values for AIC and BIC. In a next step, we’ll have to check the formulas implemented in the programs. It would be very nice if at least two of the programs produced the same values, as it makes comparisons easier for beginners.
The OpenMX beta version 0.1.2-708 is now released to invited beta testers.
A team from the University of Virginia has started to port the classic mx program to R and provide us with a powerful new SEM tool:
OpenMx is free and open source software for use with R that allows estimation of a wide variety of advanced multivariate statistical models.
The closed beta started yesterday and will open in a few months.
This is certainly great news for the SEM community and I look forward to trying it. The only thing that bugs me in all free SEM programs, including mx and John Fox’s sem package for R, is the verbose and ugly syntax for specifying models. Some oldschool LISREL users might find matrix notation easy to use, but the rest of us have been spoilt by the wonderfully concise Mplus. Jens and I did at one point plan to implement Mplus syntax on top of the sem package but I bet the OpenMx folks will beat us there!
See the excellent OpenMx homepage for details (and please vote for Mplus parsing!!)
Today, SEMNET contained the following posting:
The PLSPM package is a new R package developed by Gaston Sanchez in collaboration with Laura Trinchera to perform standard PLS analyses (PLS-PM, PLS Regression and NIPALS) and some extensions such as multigroup comparison and clustering via REBUS-PLS algorithm. On the CRAN web site you can find the last version of this package as well as tutorials. For more information you can contact Gaston (Gaston.Sanchez@upc.edu) or Laura (ltrinche@unina.it).
Thus, happy news for all fellows enjyoing PLS modeling with R!
If you enjoyed the post, make sure you subscribe to the RSS feed.
The SEM Blog by Jens Vogelgesang and Michael Scharkow.
Powered by WordPress. Design by 