Introduction of kernel functions in PLS can significantly improve the explanatory power of regression. Hope it comes true in future.

MARS is very good for modeling where the variables have a change in importance across a data set. fx=f(x) if a> some # and 0 otherwisesee https://en.wikipedia.org/wiki/Kriging#:~:text=From%20Wikipedia%2C%20the%20free%20encyclopedia,process%20g...

Right now XLstat offers only an annual subscription. I have the need to use XLstat and prefer not to use other software, but with a single, small project cannot justify $1395 for a couple days worth of analyses. I'd gladly pay for a month's subscr...

Allow Likert-type scale data to be used to automatically generate k*k pairwise comparisons, then populate a k x k square by summing the columns, to populate a k x k square matrix, and take the geometric mean of each row, then normalize the geometr...

I think will be very interesting if the NIPALS computation will be possible in all missing data selection

In a linear hierarchy regression, it is important to know the change in R-squared to know if additional variables in a model increase the association with the predictor variables between the models (independent variables).

Would be nice to be able to integrate the area under the curve in linear and non-linear models.

I think XLSTAT provides partial least squares models but not general least squares. This would be very practical for time series analysis, for instance.

In the procedure Similarity /Dissimilarity matrices but also in Cluster Analysis it is possible to use a variety of similarity and dissimilarity measures. Choosing the right one is most important. To understand all these measures it would be han...

When performing the Kruskal-Wallis analysis of variance on k samples, up until now, you can choose between Dunn, Conover-Iman and Steel-Dwass-Critchlow-Fligner with the addition of Bonferroni correction (for the first two mentioned, respectively) ...