Hi all,
The new version of STAR (0.2-2) is now available on CRAN.
* An error in function varianceTime has been corrected (thanks to Chong Gu).
* The vignette has been updated.
* A new vignette is available on STAR web sire:
http://sites.google.com/site/spiketrainanalysiswithr/
* The new goodness-of-fit test for spike train models announced in the
previous release in now fully functional thanks to feed-back from Vilmos
Prokaj. The test requires the computation of the parameters of a tight
confidence region within which a standard Wiener process "stays" in,
say,
95% of its realizations. The tightest region was shown by Kendall et al
(2007, Statistics and Computing, 17, 1) to have boundaries well approximated
by: a + b*sqrt(t) (where a and b have to be adjusted to get the correct
coverage probability). These 2 parameters are found using the method of
Loader and Deely (J Statist Comput Simulation, 27, 95). The latter method is
also implemented for arbitrary boundaries. The tight case uses C code while
the general case uses pure R code (which is reasonably fast). Error bounds
are also computed.
* Utility functions have been added to estimate the conditional intensity
(CI) of spike trains with smoothing spline (then the hard work is done by
the gss package).
* Detailed demos on how to estimate the CI are now added to the STAR web
site: http://sites.google.com/site/spiketrainanalysiswithr/
A happy new yeaR to all of you,
Christophe
--
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.
--
Christophe Pouzat
Laboratoire de Physiologie Cerebrale
CNRS UMR 8118
UFR biomedicale de l'Universite Paris-Descartes
45, rue des Saints Peres
75006 PARIS
France
tel: +33 (0)1 42 86 38 28
fax: +33 (0)1 42 86 38 30
mobile: +33 (0)6 62 94 10 34
web: http://www.biomedicale.univ-paris5.fr/physcerv/C_Pouzat.html
sip:christophe.pouzat at ekiga.net <sip%3Achristophe.pouzat at ekiga.net>
[[alternative HTML version deleted]]
_______________________________________________
R-packages mailing list
R-packages at r-project.org
https://stat.ethz.ch/mailman/listinfo/r-packages
