XLSTAT-PLS
XLSTAT-PLS is an Excel add-in that has been developed to provide XLSTAT-Pro users with a powerful solution for Partial Least Squares (PLS) regression. All XLSTAT-PLS functions have been intensively tested against other software to guarantee the users fully reliable results.
Partial Least Squares Regression is a method that has been developed during the 80s, and that is now used in more and more industries and research fields. It frees oneself from some of the constraints of the classical linear regression (Ordinary Least Squares regression - OLS): with PLS regression it is possible to model one or more dependent variables by a very high number of explanatory variables whatever the number of observations without risking to obtain an overfitted model. This module also makes available PCR (Principal Components Regression and OLS regression.
FEATURES:
PLS REGRESSION:
- PLS1 and PLS2 regression
- Explanatory variables can be quantitative and/or qualitative
- Computes and displays components
- Correlation charts and biplots
- Equations of the models
- Standardized coefficients and confidence intervals
- Predictions and residuals
- Predictions and residuals on validation set
- Predictions on prediction set
Tutorial1
PCR REGRESSION:
- Explanatory variables can be quantitative and/or qualitative
- PCA with options adapted for the regression
- Correlation and observations charts and biplots
- Goodness of fit statistics, analysis of variance
- Model coefficients with the components
- Model coefficients with the input variables
- Standardized coefficients and confidence intervals
- Display of the equation
- Predictions and residuals
- Predictions and residuals on validation set
- Predictions on prediction set
OLS REGRESSION:
- Explanatory variables can be quantitative and/or qualitative
- Four methods for selecting variables with several possible criteria
- Analysis of variance, Type I SS, Type III SS
- Model coefficients
- Standardized coefficients and confidence intervals
- Display of the equation
- Predictions and residuals
- Predictions and residuals on validation set
- Predictions on prediction set