nextdiff                 package:sca                 R Documentation

_C_o_m_p_u_t_e _t_h_e _N_e_x_t _S_i_m_p_l_e _D_i_f_f_e_r_e_n_c_e-_C_o_m_p_o_n_e_n_t _f_o_r _S_C_A

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

     Compute the next simple difference-component; this is an auxiliary
     function for 'sca'.

_U_s_a_g_e:

     nextdiff(S, P, withinblock, criterion)

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

       S: correlation/covariance matrix

       P: component matrix

withinblock: logical indicating whether any given difference-component
          should only involve variables belonging to the same
          block-component.

criterion: character string specifying the optimality criterion, see
          'sccrit' for details.

_D_e_t_a_i_l_s:

     Uses 'firstpcres(S,P)' and subsequently 'shrinkdiff()', the latter
     in a loop when 'withinblock' is true.

     In order to ensure uniqueness, we ensure that the first (non zero)
     entry of the principal component is always _positive_.

_V_a_l_u_e:

     a list with components 

       P: the new component matrix, i.e. the input 'P' with one new
          column appended.

  nextpc: the next principal component with many entries set to 0.

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

     Valentin Rousson rousson@ifspm.unizh.ch and Martin Maechler
     maechler@stat.math.ethz.ch.

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

     'shrinkdiff'; 'sca', also for references

