here is the hidden area text

The new features in version 2.1-2 are as follows:

package gamlss:

  • The function histSmo() is added for density estimation.
  • The function histDist() now has the function gamlssML() as its main fitting function. The fitting function  gamlss() is only used if gamlssML() fails.
  • The function gamlssML() has now an argument start.from.
  • In the function fitDist(), the normal distribution NO() is added to the list of “.realline” so it also appears in the   AIC list
  • The  function gamlssML() has now method summary()
  • A bugs is corrected in the function package vcov(), thanks to Tom Jagger
  • The function random() is modified to allow Local maximum likelihood estimation of the smoothing parameter lambda
  • The function pvc() has been modified to allow fixing the dfs when the “by” argument is a factor (Tim Cole suggest it).
  • The predict.gamlss() function works now with offsets
  • The worm plot function wp() works now with any fitted object which has the method resid()
  • The function dop() is renamed as dtop() and now works with any fitted object which has the the method resid()
  • The function fitted.plot() is renamed fittedPlot() to avoid S3 problems
  • pvc(): now predict is working when the argument  “by” is a continuous variable, thanks to Torsten Hothorn for point out to us. The fitted function fitDist() is introduced for fitting different distributions to a single set of data. This function fits several parametric distributions to a vector of data and chooses the one with minimum GAIC.

package gamlss.dist:

  • The distributions SHASHo and SHASHo2 and PARETO2o are added to the package gamlss.dist
  • The random generating function rDEL() is now corrected thanks to Dr. Conrad Burden
  • The following distributions are added to package gamlss.dist: YULE, WARING, GEOM, IGAMMA, PARETO2
  • The distributions BCTo, BCPEo, and BBCGo are added so BCT, BCPE and BCCG can have”log” as a default

package gamlss.util:

  • The functions fitFixBP() and fitFreeKnots() for fitting fixed and free break points respectively have been improved

2 Responses to “Version 4.1-2”

  1. Dear
    Recently, I meet a problem about use of centile.pred() in GAMLSS.
    1) Water level data from 1954 to 2000 to build a model is calibrated as following
    H<-gamlss(y~r1+t1,,family=GA, data=water)
    where, r1 and t1 are the two explanatory variables.
    2)And then, I want to predict water level during 2001–2010 using related predictors (r1 and t1) in terms of the 5th, median and 95th percentiles based on the another function.
    centiles.pred(H, xname='?',xvalues='year') .
    What should I use in the question mark? Or Is any other solution for this problem.
    Thanks a lot.


Leave a Reply

XHTML: You can use these tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>