Title
somersd_mata -- Mata functions used by somersd
Contents
Online help entry Function Purpose ---------------------------------------------------------------------------
blncdtree() blncdtree() create balanced binary search tree _blncdtree() create balanced binary search subtre > e tidottree() concordance-discordance difference t > otals tidot() concordance-discordance difference t > otals (inefficiently)
_u2jackpseud() _u2jackpseud() jackknife pseudovalues for degree-2 > U statistics _v2jackpseud() jackknife pseudovalues for degree-2 > V statistics
_somdtransf() _somdtransf() transformations used by somersd
_bcsf_bracketing() _bcsf_bracketing() bracketing a value of a monotonic bo > unded step function _bcsf_bisect() bracket convergence by bisection met > hod _bcsf_regula() bracket convergence by regula falsi > method _bcsf_ridders() bracket convergence by Ridders' meth > od
---------------------------------------------------------------------------
Description
The functions above are used by somersd to calculate jackknife pseudovalues for sample means, Hoeffding U statistics and Von Mises V statistics, and by censlope to carry out bracket convergence in the iterative estimation of percentile slopes. The pseudovalues are based on concordance-discordance difference totals, which can be calculated efficiently using balanced binary search trees. The program somersd uses these pseudovalues to calculate delta-jackknife confidence intervals for Somers' D or for Kendall's tau-a.
Author
Roger Newson, Imperial College London, UK. Email: r.newson@imperial.ac.uk
Also see
Online: mata, somersd, if installed