

   PPootttthhooffff aanndd RRooyy ggrroowwtthh ccuurrvvee mmooddeell

        potthoff(response, x, ccov=NULL, times=NULL, torder=0, orthogonal=T)

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

   response: A matrix or dataframe of response values.

          x: A matrix defining the complete intersubject dif-
             ferences.

       ccov: A matrix of columns of the baseline covariates to
             be actually fitted, with one row per individual.

      times: A vector of unequally spaced times when they are
             the same for all individuals. Not necessary if
             equally spaced.

     torder: Order of the polynomial in time to be fitted. If
             non-numeric, the full model in time is fitted.

   orthogonal: If TRUE, uses orthogonal polynomials for time,
             otherwise only centres times at their mean.

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

        `potthoff' fits the Potthoff and Roy repeated measure-
        ments growth curve model with unstructured covariance
        matrix to completely balanced data.

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

        A list of class potthoff is returned.

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

        J.K. Lindsey

   SSeeee AAllssoo::

        `carma', `elliptic', `lm'.

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

        y <- matrix(rnorm(40),ncol=5)
        x <- gl(2,4)
        # full model with treatment effect
        potthoff(y, ~x, torder="f", ccov=~x)
        # no time trend with treatment effect
        potthoff(y, ~x, torder=0, ccov=~x)
        # quadratic time with treatment effect
        potthoff(y, ~x, torder=2, ccov=~x)
        # full model without treatment effect
        potthoff(y, ~x, torder="f")
        # linear time without treatment effect
        potthoff(y, ~x, torder=1)

