Announcing ASSIST

[Yuedong Wang]

We have put together a package ASSIST, a collection of
R functions for fitting many spline-based 
non-parametric/semi-parametric linear/non-linear
fixed/mixed models. The general form of smoothing splines 
based on reproducing kernel Hilbert spaces is used to 
model nonparametric functions. Some well known special 
cases include polynomial, periodic, spherical, thin-plate 
splines and L splines as well as smoothing spline ANOVA 
models. It also covers some cases of generalized additive 
models,  multiple index models, varying coefficient models, 
functional linear models, and self-modeling nonlinear 
regression models. Users may also provide their own kernels.
 
The package contains five main R functions:
 
1. ssr fits non-parametric regression models for
   independent and correlated Gaussian data, and for
   independent binomial, Poisson and Gamma data;
2. slm fits semi-parametric linear mixed-effects models;
3. nnr fits non-parametric nonlinear regression models;
4. snr fits semi-parametric nonlinear regression models; 
5. snm fits semi-parametric nonlinear mixed-effects models.
 
These extend such existing functions in R (S+) as
smooth.spline, gam, nls, lme and nlme. The current version can 
deal with small to moderate data sets only. We are working on 
options to deal with large data sets.
 
ASSIST may be obtained at CRAN 

http://cran.us.r-project.org/
 
  or

http://www.pstat.ucsb.edu/faculty/yuedong/software
 
Detailed information about these functions and many
examples can be found in the manual of this package,
which again can be downloaded from the above site.
 
Comments and suggestions are highly appreciated!

Yuedong Wang and Chunlei Ke