

   GGuueessss tthhee hhoorriizzoonnttaall aassyymmppttoottee oonn tthhee rriigghhtt ssiiddee

        NLSstRtAsymptote(xy)

   AArrgguummeennttss::

         xy: a `sortedXyData' object

   DDeessccrriippttiioonn::

        This function provides an initial guess at the horizon-
        tal asymptote on the right side (larger values of `x')
        of the graph of `y' versus `x' from the `xy' object.
        It is mostly used within the `initial' functions for a
        self-starting nonlinear regression models, which are in
        the `selfStart' class.

   VVaalluuee::

        A single numeric value which is a guess at the `y'
        value that would be the asymptote for large `x'.

   AAuutthhoorr((ss))::

        Jose Pinheiro and Douglas Bates

   EExxaammpplleess::

        library( lme )
        data( DNase )
        DNase.2 <- DNase[ DNase$Run == "2", ]
        DN.srt <- sortedXyData( expression(log(conc)), expression(density), DNase.2 )
        NLSstRtAsymptote( DN.srt )

