WGA Algorithm

WGA =   \begin{displaymath} \frac{d_b}{k_1d_{1}+k_2d_{2}+\alpha} \end{displaymath}


  • description of letters
    • $n_i$ is the total number of items in group i.
    • $t_i$ is the number of pairings within a group.
    • $d_i$ is the average distance among all sample pairs within the group.
    •  \begin{displaymath} \alpha \end{displaymath} is used to make sure that division by zero never happens.
  • detailed formula
    • $d_b = | mean_1 - mean_2 |$
    •  \begin{displaymath} t_i = (n_i-1)+(n_i-2)+...+(n_i-(n_i-1)) = \frac{n_i(n_i-1)}{2} \end{displaymath}
      •  \begin{displaymath} k_1=\frac{t_1}{t_1+t_2} \end{displaymath}
      •  \begin{displaymath} k_2=\frac{t_2}{t_1+t_2} \end{displaymath}
    • $d_i =  \frac{|v_1 - v_2| + |v_1 - v_3| + ... + |v_1 - v_{n_i}| + |v_2 - v_3| + ... + |v_2 - v_{n_i}| + ... + |v_{n_{i-1}} - v_{n_i}|}{t_i} $
    •  \begin{displaymath} \alpha \end{displaymath}
      • if $k_1d_{1}+k_2d_{2}$ == 0
        • if $ d_{b}$ == 0 then WGA = 0
        • else  \begin{displaymath} \alpha \end{displaymath} = $\frac {d_{b}} {1000}$
      • else  \begin{displaymath} \alpha \end{displaymath} = 0

WGA documentations

Edit | Attach | Print version | History: r20 | r15 < r14 < r13 < r12 | Backlinks | View wiki text | Edit WikiText | More topic actions...
Topic revision: r14 - 04 Apr 2006, JoanZhang
 

This site is powered by FoswikiCopyright © 2013-2022 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Vanderbilt Biostatistics Wiki? Send feedback