[R-pkgs] STAR (Spike Train Analysis with R) uploaded on CRAN
christophe.pouzat at gmail.com
Wed Sep 10 10:56:06 CEST 2008
I've uploaded STAR (Spike Train Analysis with R) on CRAN two days ago.
The package is designed to analyze neuronal spike (action potential)
trains. It uses S3 classes and methods and makes heavy use of other
CRAN packages like gss, R2HTML, mgcv, survival.
* Analysis of both spontaneous and stimulus evoked activity is
implemented for single neuron spike trains as well as for many neurons
* Several bi variate duration distribution can be automatically fitted
to "spontaneous" spike trains. Tests for renewal processes are
* A non-parametric, smoothing spline based, approach to the estimation
of the conditional intensity of the trains is also available (this
feature is under development).
* The full collection of Ogata's tests (1988, JASA, 83:9) for
point-process models is implemented and the data set used by him
(table 1 of the paper) is made available.
* A new goodness of fit test for point-process is also proposed based
on the fact that the martingale of the process plotted as a function
of the integrated conditional intensity starts behaving very quickly
as a standard Wiener process (if the model is correct). By the way I
have no clean and formal proof of the latter statement so anyone able
(and wiling) to give me some help on that is invited to send me an
* Many data sets are included as well as vignette (covering only part
of the present functionalities) and several demos.
Any feed-back, comment, critic, is welcome.
A Master Carpenter has many tools and is expert with most of them. If you
only know how to use a hammer, every problem starts to look like a nail.
Stay away from that trap.
Richard B Johnson.
Betweeen May and October I'll be mainly staying
in the Peter Kloppenburg's lab in Cologne, You can
call me there at: +49 (0)221 470 5207
Laboratoire de Physiologie Cerebrale
CNRS UMR 8118
UFR biomedicale de l'Universite Paris-Descartes
45, rue des Saints Peres
tel: +33 (0)1 42 86 38 28
fax: +33 (0)1 42 86 38 30
mobile: +33 (0)6 62 94 10 34
sip:christophe.pouzat at ekiga.net
More information about the R-packages