[Forum SIS] PLS (Partial Least Squares) News: XLSTAT-PLSPM software + Scientific Award + PLS Handbook on YouTube

[Vincenzo Esposito Vinzi]

Dear PLS researchers and users,

I am pleased to announce the most recent news related to the world of PLS (Partial Least Squares):

1) New features of XLSTAT-PLSPM software - JUST RELEASED!
2) Establishment of a Scientific PLS Award (1,000 Euros)
3) YouTube video related to PLS Handbook

Enjoy these news!

Vincenzo Esposito Vinzi


XLSTAT-PLSPM software: New Features in version 2010.4 - JUST RELEASED!
------------------------------------------------------------------------------------------------------------

NEW and most advanced features of the XLSTAT module for PLS Path Modeling (XLSTAT-PLSPM, version 2010.4),
the only software that allows using the PLS approach (both PLS Path modeling and PLS regression) without leaving Microsoft Excel:

TREATMENT OF MISSING DATA:
- The NIPALS algorithm
- The PLS regression algorithm
- When asking latent variables scores in the original scale (normalized weights): ipsative means and renormalization

BOOTSTRAP MODULE:
- Confidence intervals for indirect effects between latent variables (very useful for testing mediating effects)
- Display of standardized loadings for each bootstrap resample
- Display of path coefficients for each bootstrap resample

PLS REGRESSION MODULE (very useful for dealing with multidimensionality, multicollinearity, missing data, scarceness of samples as compared to variables):
- Full-featured PLS regression module for the estimation of:
Outer weights for multidimensional (formative) blocks
Path coefficients in presence of multicollinearity between latent variables
- Outliers analysis

STRUCTURAL MODEL:
- Effect size (f2) for explanatory latent variables in structural equations
- PLS Confirmatory Analysis (very useful for multi-block data analysis)
- Nonlinear relationships between latent variable scores via the non linear regression options commonly available "Modeling data" module of XLSTAT.

MEASUREMENT MODEL:
- Loadings (unstandardized) when working with raw manifest variables
- Location parameters when working with raw manifest variables
- Estimation of higher order dimensions via external deflation (very useful for multi-block data analysis)
- Possibility of choosing a different number of dimensions per each block

UNCOVERING LATENT SEGMENTS:
- REBUS (REsponse-based Unit Segmentation) approach with local models
- Validation of REBUS results via t-tests and permutation tests in multi-group analysis

A detailed description of the XLSTAT-PLSPM product is available at:
http://www.xlstat.com/en/products/xlstat-plspm/

A 30-days free trial version can be downloaded from the XLSTAT website:
http://www.xlstat.com/en/download/

Several tutorials are available at:
http://www.xlstat.com/en/support/tutorials/

For any further information, please refer to Mr. Thierry Fahmy (fahmy at xlstat.com).


SCIENTIFIC PLS AWARD
-----------------------------------

After contributing to the first PLS Award in 2009, attributed to Michel Tenenhaus for his major contributions to the PLS methods (Regression and Path Modeling),
Addinsoft announces an annual award, the PLS Award, endowed with 1,000 Euros. The next prize will be awarded in 2011 by a committee headed by
Michel Tenenhaus to reward a major breakthrough in the field of PLS methods.


YouTube VIDEO - HANDBOOK OF PLS
------------------------------------------------------

The "Handbook of Partial Least Squares: Concepts, Methods and Applications" has been recently released by Springer.
A video explaining the content of this Handbook is available on line.

You may access this video either via YouTube at:
http://www.youtube.com/watch?v=WihfjntEWzo

or via the XLSTAT News website at:
http://www.xlstat.com/fr/company/news/2010.htm#pls_workshop




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.stat.unipg.it/pipermail/sis/attachments/20100704/aa09c33c/attachment.html>