gdist                 package:mvpart                 R Documentation

_D_i_s_s_i_m_i_l_a_r_i_t_y _M_e_a_s_u_r_e_s

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

     The function computes useful dissimilarity indices which are known
     to have a good rank-order relation with gradient separation and
     are thus efficient in community ordination with multidimensional
     scaling.

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

     gdist(x, method="bray", keepdiag=FALSE, full=FALSE, sq=FALSE)

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

       x: Data matrix

  method: Dissimilarity index 

keepdiag: Compute amd keep diagonals

    full: Return the square dissimilarity matrix

      sq: Square the dissimilarities - useful for distance-based
          partitioning

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

     The function knows the following dissimilarity indices:

       'euclidean'   d[jk] = sqrt(sum (x[ij]-x[ik])^2)
       'manhattan'   d[jk] = sum(abs(x[ij] - x[ik]))
       'gower'       d[jk] = sum (abs(x[ij]-x[ik])/(max(i)-min(i))
       'canberra'    d[jk] = (1/NZ) sum ((x[ij]-x[ik])/(x[ij]+x[ik]))
       'bray'        d[jk] = (sum abs(x[ij]-x[ik])/(sum (x[ij]+x[ik]))
       'kulczynski'  d[jk] 1 - 0.5*((sum min(x[ij],x[ik])/(sum x[ij]) + (sum min(x[ij],x[ik])/(sum x[ik]))
       'maximum'     d[jk] = max(abs(x[ij] - x[ik]))
       'binary'      d[jk] = max(abs(x[ij]>0 - x[ik]>0))
       'chord'       d[jk] = sqrt((sum (x[ij]-x[ik])^2)/(sum (x[ij]+x[ik])^2))

     where NZ is the number of non-zero entries.

     Infamous ''double zeros'' are removed in Canberra dissimilarity.

     Euclidean and Manhattan dissimilarities are not good in gradient
     separation without proper standardization but are still included
     for comparison and special needs.

     Some of indices become identical or rank-order similar after some
     standardizations.

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

     Should be interchangeable with 'dist' and returns a distance
     object of the same type.

_N_o_t_e:

     The  function is an alternative to 'dist' adding some ecologically
     meaningful indices.  Both methods should produce similar types of
     objects which can be interchanged in any method accepting either. 
     Manhattan and Euclidean dissimilarities should be identical in
     both methods, and Canberra dissimilary may be similar.

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

     Jari Oksanen  - modified Glenn De'ath (Dec 03)

_R_e_f_e_r_e_n_c_e_s:

     Faith, D.P, Minchin, P.R. and Belbin, L. (1987) Compositional
     dissimilarity as a robust measure of ecological distance.
     _Vegetatio_ 69, 57-68.

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

     data(spider)
     spider.dist <- gdist(spider[1:12,])

