lmeLevel-class             package:lme4             R Documentation

_C_l_a_s_s "_l_m_e_L_e_v_e_l"

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

     An object representing a single level of random effects in a
     linear mixed-effects model.

_O_b_j_e_c_t_s _f_r_o_m _t_h_e _C_l_a_s_s:

     Objects can be created by calls of the form `new("lmeLevel", ...)'
     or by calls to the generic constructor `lmeLevel'.

_S_l_o_t_s:

     `_p_r_e_c_i_s_i_o_n': A precision matrix of class `"pdMat"'

     `_g_r_o_u_p_s': The grouping factor (of class `"factor"') for the level.

     `_c_o_l_u_m_n_s': An `"integer"' vector of the columns in the model
          matrix that correspond to this level.

     `_p_a_r_s_I_n_d': An `"integer"' vector giving the subscripts in the
          overall parameter vector that apply to this level.

     `_o_r_i_g_i_n_a_l_R_o_w_s': A list of integer vectors of row numbers in the
          original model matrix that correspond to each of the levels
          of the grouping factor.

     `_d_e_c_o_m_p_o_s_e_d_R_o_w_s': A list of integer vectors of row numbers in the
          decomposed model matrix that correspond to each of the levels
          of the grouping factor.

     `_s_t_o_r_e_d_R_o_w_s': A list of integer vectors of row numbers in the
          stored data matrix that correspond to each of the levels of
          the grouping factor.

     `_n_r_o_w': `integer': the number of rows in the precision matrix.

     `_u_p_d_a_t_e_F_a_c_t_o_r': Object of class `"matrix"' - the matrix `A' used
          to evaluate the gradient of the log-likelihood and to
          formulate the ECME step.

     `_h_e_s_s_i_a_n_A_r_r_a_y': Object of class `"array"' - the array `H' used to
          evaluate the Hessian of the log-likelihood.

     `_n_l_e_v': `integer': the number of levels in the grouping factor

_M_e_t_h_o_d_s:

     _E_M_u_p_d_a_t_e<- `signature(x = "lmeLevel", nlev = "missing", value =
          "matrix")': update the level within the EM or ECME algorithm

     _L_M_E_g_r_a_d_i_e_n_t `signature(x = "lmeLevel", A = "matrix", nlev =
          "missing")': calculate the gradient of the log-likelihood
          with respect to the parameters from the decomposition of the
          scaled BLUPs and their variances.

     _c_o_e_f `signature(object = "lmeLevel")': extract the parameters
          (from the precision matrix)

     _c_o_e_f<- `signature(object = "lmeLevel", value = "numeric")': set
          the parameters.

_N_o_t_e:

     In `reStruct' objects the random-effects terms and the
     fixed-effects terms and the response are all represented by
     `lmeLevel' objects.  The `precision' slot in the fixed-effects and
     response objects is an uninitialized `pdMat' objects and the
     `nrow' slot is zero.

_A_u_t_h_o_r(_s):

     Saikat DebRoy saikat@stat.wisc.edu and Douglas Bates
     bates@stat.wisc.edu

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

     `lmeLevel'

