Variable Selection Based on Direction Vectors

  • Kyungmee Choi (Asistant Professor, Department of mathematics, Hongik University)
  • Published : 1998.04.01


We review a multivariate version of Kendall's tau based on direction vectors of observations. And with this statistic we propose an analog of the forward variable selection method which selects a set of independent variables for further studies to build the eventual predicting model. This method does not assume the distributions of observations and the linear model and it is strong to the outliers with high asymptotic efficiencies relative to the parametric Pearson's correlation coefficient.



  1. Journal of the American Statistical Association v.91 On a geometric notion of quanltiles for multivariate data Chaudhury, P.
  2. Nonparametric Multivariate Multisample Tests for the Location Problem and Multivariate Regression Based on Directions of Data(Unpublished Ph.D Dissertation) Choi, K.
  3. A multivariate version of Kendall's tau, a manuscript in revision Choi, K.;Marden, J.
  4. Practical Nonparametric Statistics(2nd ed.) Conover W. J.
  5. Total Biochemical Oxygen Demand of Animal Manures(Unpublished dissertation) Moore, J. A.
  6. International Statistical Review v.58 A survey of Multidimensional medians Small, C. G.
  7. Applied Linear Regression Weisberg, S.