

pdBlocked(nlme)                              R Documentation

_P_o_s_i_t_i_v_e_-_D_e_f_i_n_i_t_e _B_l_o_c_k _D_i_a_g_o_n_a_l _M_a_t_r_i_x

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

     This function is a constructor for the `pdBlocked'
     class, representing a positive-definite block-diagonal
     matrix. Each block-diagonal element of the underlying
     matrix is itself a positive-definite matrix and is rep-
     resented internally as an individual `pdMat' object.
     When `value' is `numeric(0)', a list of uninitialized
     `pdMat' objects, a list of one-sided formulas, or a
     list of vectors of character strings,  `object' is
     returned as an uninitialized `pdBlocked' object (with
     just some of its attributes and its class defined) and
     needs to have its coefficients assigned later, gener-
     ally using the `coef' or `matrix' replacement func-
     tions. If `value' is a list of  initialized `pdMat'
     objects, `object' will be constructed from the list
     obtained by applying `as.matrix' to each of the `pdMat'
     elements of `value'. Finally, if `value' is a list of
     numeric vectors, they are assumed to represent the
     unrestricted coefficients of the block-diagonal ele-
     ments of the  underlying positive-definite matrix.

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

     pdBlocked(value, form, nam, data, pdClass)

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

   value: an optional list with elements to be used as the
          `value' argument to other `pdMat' constructors.
          These include: `pdMat' objects, positive-definite
          matrices, one-sided linear formulas, vectors of
          character strings, or numeric vectors. All ele-
          ments in the list must be similar (e.g. all one-
          sided formulas, or all numeric vectors). Defaults
          to `numeric(0)', corresponding to an uninitialized
          object.

    form: an optional list of one-sided linear formulas
          specifying the row/column names for the block-
          diagonal elements of the matrix represented by
          `object'. Because factors may be present in
          `form', the formulas needs to be evaluated on a
          data.frame to resolve the names they define. This
          argument is ignored when `value' is a list of one-
          sided formulas. Defaults to `NULL'.

     nam: an optional list of vector of character strings
          specifying the row/column names for the block-
          diagonal elements of the matrix represented by
          object. Each of its components must have length
          equal to the dimension of the corresponding block-
          diagonal element and unreplicated elements. This
          argument is ignored when `value' is a list of vec-
          tor of character strings. Defaults to `NULL'.

    data: an optional data frame in which to evaluate the
          variables named in `value' and `form'. It is used
          to obtain the levels for `factors', which affect
          the dimensions and the row/column names of the
          underlying matrix. If `NULL', no attempt is made
          to obtain information on any `factors' appearing
          in the formulas. Defaults to the parent frame from
          which the function was called.

 pdClass: an optional vector of character strings naming the
          `pdMat' classes to be assigned to the individual
          blocks in the underlying matrix. If a single class
          is specified, it is used for all block-diagonal
          elements. This argument will only be used when
          `value' is missing, or its elements are not
          `pdMat' objects. Defaults to `"pdSymm"'.

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

     a `pdBlocked' object representing a positive-definite
     block-diagonal matrix, also inheriting from class
     `pdMat'.

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

     Jose Pinheiro and Douglas Bates

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

     `as.matrix.pdMat', `coef.pdMat', `matrix<-.pdMat'

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

     library(nlme)
     pd1 <- pdBlocked(list(diag(1:2), diag(c(0.1, 0.2, 0.3))),
                      nam = list(c("A","B"), c("a1", "a2", "a3")))
     pd1

