GetAlpha            package:GeneSelector            R Documentation

_H_e_l_p_e_r _f_u_n_c_t_i_o_n _f_o_r _s_t_a_b_i_l_i_t_y _a_s_s_e_s_s_e_m_e_n_t.

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

     Both GetStabilityLm and GetStabilityOverlap depend on a parameter
     'alpha' if 'decay=exponential'. If the weights are based on ranks,
     then a nonlinear regression of the form 
      'pval = 1- exp(-alpha*rank)' 
      can be used to find an appropriate value for 'alpha' via
     nonlinear least squares. In order to adjust for too 'optimistic'
     p-values, multiple testing adjustments should be used, s.
     AdjustPvalues.

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

     GetAlpha(ranking, pval, alpha0 = 0.01)

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

 ranking: A numeric vector of ranks, regarded as regressor.

    pval: A numeric vector of p-values corresponding to the vector
          'ranking'.

  alpha0: A starting value for the nonlinear least squares estimation
          procedure passed to nls

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

     The nonlinear least squares estimator for 'alpha', s. description.

_N_o_t_e:

     It is more or less  equivalent to use a p-value based ranking
     instead of ranks combined with this procedure.

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

     Martin Slawski martin.slawski@campus.lmu.de 
      Anne-Laure Boulesteix <URL: http://www.slcmsr.net/boulesteix>

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

     GetStabilityLm, GetStabilityOverlap, AdjustPvalues, nls

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

     ### rankings
     ranks <- 1:100
     ### corresponding p-values
     pvals <- 1-exp(-0.01*ranks) + rnorm(100, sd=0.001)
     ### determine alpha
     alphaopt <- GetAlpha(ranks, pvals, alpha0 = 0.01)

