Algorithm for the Constrained Chebyshev Estimation in Linear Regression

  • Kim, Bu-yong (Department of Statistics Sookmyung Women's University)
  • 발행 : 2000.04.01

초록

This article is concerned with the algorithm for the Chebyshev estimation with/without linear equality and/or inequality constraints. The algorithm employs a linear scaling transformation scheme to reduce the computational burden which is induced when the data set is quite large. The convergence of the proposed algorithm is proved. And the updating and orthogonal decomposition techniques are considered to improve the computational efficiency and numerical stability.

키워드

참고문헌

  1. Nordisk Tidskrift for Information-behandling v.17 The Discrete Linear Restricted Chebyshev Approximation Abdelmalek, N.N.
  2. Mathematical Programming v.5 On L₁ and Chebyshev Estimation Appa, G.;Smith, C.
  3. ACM Transactions on Mathematical Software v.1 Algorithm 495: Solution of an Overdetermined System of Linear Equations in the Chebyshev Norm Barrodale, I.;Phillips, C.
  4. SIAM Review v.6 Norms for Smoothting and Estimation Rice, J.R.;White, J.S.
  5. Naval Research Logistics v.35 An Assumption-Free Convergence Analysis for a Perturbation of the Scaling Algorithm for Linear Programs, with Application to the L₁Estimation Problem Sherali, H.D.;Skarpness, B.O.;Kim, B.Y.
  6. Computers and Operations Research v.10 A Linear Programming Algorithm for the Simple Model for Discrete Chebychev Curve Fitting Sklar, M.G.;Armstrong, R.D.
  7. Commun. Statist.-Simula. Computa. v.13 An Algorithm for Discrete Chebychev Curve Fitting for the Simple Model using a Dual Linear Programming Approach Sklar, M.G.;Armstrong, R.D.