ADDRESS DECLINE IN THE LEAST SQUARES PROBLEM OF A INTERIOR POINT METHOD FOR LINEAR PROGRAMMING

• Autores: Rojas Jerónimo, Jenny, De la Cruz Chávez, Carlos
• Formato: artículo
• Fecha de Publicación: 2015
• Institución: Universidad Nacional de Trujillo
• Repositorio: Revistas - Universidad Nacional de Trujillo
• Lenguaje: español
• OAI Identifier: oai:ojs.revistas.unitru.edu.pe:article/1240
• Enlace del recurso: https://revistas.unitru.edu.pe/index.php/SSMM/article/view/1240
• Nivel de acceso: acceso abierto
• Materia: Linear Programming; inner elipsoid; least square; descent direction; Programación Lineal; elipsoide interior; mínimos cuadrados; dirección de descenso

This research work solves the problem of least squares that requires inner elipsoid algorithm to determine the descent direction; giving solution to linear programming problems by means of this methodof interior points. We solve the least squares problem using auxiliary function with logarithmic barrier and an approximation of the original matrix factorization by a matrix of rank one update to nally use the Sherman-Morrison-Woodburry formula and determining the inverse of the current matrix thus solving the least squares problem and obtaining a approximation to the descent direction.