[R-sig-Epi] Epi 1.0.7 now out

[BXC (Bendix Carstensen)]

Dear R-sig-Epi list,

We have now made a major update of the Epi package, to version 1.0.7.
It has just appeared on CRAN.

For a few days the version 1.0.5 was there, but there were grave bugs
in splitLexis and cutLexis, so please update to 1.0.7 if you have the
1.0.5. ( Use: installed.packages()["Epi","Version"] to check ).

The biggest change since the previous version 0.7.0 on CRAN is the
introduction of a new machinery for representation of follow-up data
through the Lexis() function that creates a Lexis object. It enables
simultaneous representation of follow-up on different timescales and
also representation of multistate data, i.e. transitions between
several different states (not only alive -> dead). Time-splitting of
follow-up data is done with the functions splitLexis(), cutLexis() and
countLexis(). Overview of multistate data is given by tab.Lexis().

Included in the package is a vignette explaining this machinery in
detail --- start the html-help (write help.start() at the command
prompt) and at the front page for the Epi package, click on a link in
"Read overview or browse directory".

Note that the old Lexis()-function (by David Clayton) is dead. It is
still in the package, but now under the name W.Lexis(). Its
functionality, splitting follow-up time, is taken over by
splitLexis().

The function epi.eff() is replaced by effx() and effx.match().

Enjoy!

Best wishes,
Bendix Carstensen,
maintainer of Epi
______________________________________________

Bendix Carstensen
Senior Statistician
Steno Diabetes Center
Niels Steensens Vej 2-4
DK-2820 Gentofte
Denmark
+45 44 43 87 38 (direct)
+45 30 75 87 38 (mobile)
bxc at steno.dk   http://www.biostat.ku.dk/~bxc