

varIdent(nlme)                               R Documentation

_C_o_n_s_t_a_n_t _V_a_r_i_a_n_c_e _F_u_n_c_t_i_o_n

_D_e_s_c_r_i_p_t_i_o_n_:

     This function is a constructor for the `varIdent'
     class, representing a constant variance function struc-
     ture. If no grouping factor is present in `form', the
     variance function is constant and equal to one, and no
     coefficients required to represent it. When `form'
     includes a grouping factor with M > 1 levels, the vari-
     ance function allows M different variances, one  for
     each level of the factor. For identifiability reasons,
     the coefficients of the variance function represent the
     ratios between the variances and a reference variance
     (corresponding to a reference group level). Therefore,
     only M-1 coefficients are needed to represent the vari-
     ance function. By default, if the elements in `value'
     are unnamed, the first group level is taken as the ref-
     erence level.

_U_s_a_g_e_:

     varIdent(value, form, fixed)

_A_r_g_u_m_e_n_t_s_:

   value: an optional numeric vector, or list of numeric
          values, with the variance function coefficients.
          If no grouping factor is present in `form', this
          argument is ignored, as the resulting variance
          function contains no coefficients. If `value' has
          length one, its value is repeated for all coeffi-
          cients in the variance function. If `value' has
          length greater than one, it must have length equal
          to the number of grouping levels minus one and
          names which identify its elements to the levels of
          the grouping factor. Only positive values are
          allowed for this argument. Default is
          `numeric(0)', which results in a vector of zeros
          of appropriate length being assigned to the coef-
          ficients when `object' is initialized (correspond-
          ing to constant variance equal to one).

    form: an optional one-sided formula of the form `~ v',
          or `~ v | g', specifying a variance covariate `v'
          and, optionally, a grouping factor `g' for the
          coefficients. The variance covariate is ignored in
          this variance function. When a grouping factor is
          present in `form', a different coefficient value
          is used for each of its levels less one reference
          level (see description section below). Several
          grouping variables may be simultaneously speci-
          fied, separated by the `*' operator, like in `~ v
          | g1 * g2 * g3'. In this case, the levels of each
          grouping variable are pasted together and the
          resulting factor is used to group the observa-
          tions. Defaults to `~ 1'.

   fixed: an optional numeric vector, or list of numeric
          values, specifying the values at which some or all
          of the  coefficients in the variance function
          should be fixed. It must have names identifying
          which coefficients are to be fixed. Coefficients
          included in `fixed' are not allowed to vary during
          the optimization of an objective function.
          Defaults to `NULL', corresponding to no fixed
          coefficients.

_V_a_l_u_e_:

     a `varIdent' object representing a constant variance
     function structure, also inheriting from class `var-
     Func'.

_A_u_t_h_o_r_(_s_)_:

     Jose Pinheiro and Douglas Bates

_S_e_e _A_l_s_o_:

     `varWeights.varFunc', `coef.varIdent'

_E_x_a_m_p_l_e_s_:

     library(nlme)
     vf1 <- varIdent(c(Female = 0.5), form = ~ 1 | Sex)

