[R] version 0.5 of my libraries
maechler at stat.math.ethz.ch
Mon Jul 5 17:10:30 CEST 1999
>>>>> On Thu, 1 Jul 1999 12:34, Jim Lindsey <jlindsey at alpha.luc.ac.be> said:
Jim> I have just uploaded version 0.5 of my libraries to the CRAN site.
Jim> Hopefully, they will shortly become available.
Jim> New features include:
Jim> 1. A function (binnest in repeated) for binary data with two levels of
Jim> nesting based on Tom Ten Have's recent Stat in Medicine paper.
Jim> (Tom's Fortran90 routines have a builtin Newton Raphson-type
Jim> optimizer with analytic first derivatives for these models.
Jim> Unfortunately, this does not appear to work nearly as well as the
Jim> Dennis and Schnabel routine in nlm that calculates numerical
Jim> derivatives. However, as it was part of the package, I left it as
And that part is not Fortran77,
e.g. Solaris' f77 (which is said to have good properties)
cannot compile it.
1) End-line-comments starting with "!" work,
but not when the "!" comes too early in an otherwise empty line.
Note however that endline "!" may not work at all on other f77's,
since afaik these are newer than Fortran77.
2) `<=' instead of `.le.' are not liked at all
[I'll send diffs to Jim privately]
Jim> 2. The elliptic function in growth now fits multivariate Student t
Jim> models as well as multivariate normal and multivariate power
Jim> exponential (all are multivariate elliptical distributions).
Jim> 3. elliptic now provides recursive predicted values for many models.
Jim> 4. many functions can now have common parameters in the location
Jim> regression and in the dispersion regression models.
Jim> 5. Plus many enhancements and a few bug corrections (see changes.txt).
Thank you !
Jim> Bug reports are welcome. Alpha version 0.6 will shortly start to
Jim> appear on www.luc.ac.be/~jlindsey/rcode.html.
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !) To: r-help-request at stat.math.ethz.ch
More information about the R-help