- npfda :
*NonParametric Functional Data Analysis*

This website proposes R and S-plus packages allowing to implement nonparametric methods for functional data (prediction, time series analysis, discrimination, classification). Functional datasets and examples of use are also available, which makes easy their implementation.

Performed by*Frédéric Ferraty*

*F. Ferraty, P. Vieu (2006). NonParametric Functional Data analysis: Theory and Practice. Springer, New York*

- salsa :
*Smoothing Alternating with Least Squares Algorithm*

The file salsa.shar (170kbytes) contains Splus functions, programs C and Fortran, allowing to do curvilinear PCA (non linear PCA) via the algorithm SALSA. For more details see "salsa.ps", a poscript file describing the model, the estimation procedure and the algorithm.

Written by*Frédéric Ferraty*.

*Besse, P. C. and F. Ferraty (1995). A fixed effect curvilinear model. Computational Statistics, 10, p 339-351.*

*Ferraty, F. (1997). Estimations de transformations optimales en ACP curvilineaire. Revue de Statistique Appliquee, XLV (1), p 5-39.*

- simreg :
*Simultaneous Nonparametric Regression*

The file regsim.shar (100kbytes) contains functions Splus, C and Fortran programs allowing to realize a simultaneous nonparametric regression of longitudinal data leading to a new form of functional PCA taking into account constraints on dimension and smoothing.

Written by*Frédéric Ferraty*.

*Besse, P. C., Cardot, H. and F. Ferraty (1997). Simultaneous non-parametric regressions of unbalanced longitudinal data. Computational Statistics & Data Analysis, 24-3, p255-270.*

- pgsplus.spl :
*Spline Estimators for the Functional Linear Model*

The Splus function pgsplus.spl allows to estimate the functional coefficient in the functional linear model when the explanatory variable is a curve and the response a scalar.

Written by*Hervé Cardot*.

*Cardot, H., Ferraty, F. and Sarda, P. (2003). Spline estimators for the functional linear model. Statistica Sinica, 13, p 571-591.*

- COMPAR :
*Tool for comparing samples using the comparison of covariance or correlation matrices*

This program gives tests on covariances or correlations matrices.

Written by*Sylvie Viguier-Pla*.

*Flury, B.N. (1988). Common Principal Components and Related Multivariate Models. Wiley, New-York*

Viguier-Pla, S. (1994). Factor-based comparison of several populations using COMPAR software. Computational Statistics, 9, p 135-163.

Viguier-Pla, S. (2004). Factor-based comparison of k populations. Statistics, 38, p 1-15.