Abstract
This article describes an approach known as the conjugate Gram-Schmidt method for estimating gradients and Hessian using function evaluations and difference quotients, and uses the Gram-Schmidt conjugate direction algorithm to minimize functions and compares it to other techniques for solving ∇f = 0. Comparable minimization algorithms are also used to demonstrate convergence rates using quotient and root convergence factors, as described by Ortega and Rheinbolt to determine the optimal minimization technique to obtain results similar to the Newton method, between the Gram-Schmidt approach and other minimizing approaches. A survey of the existing literature in order to compare Hestenes’ Gram-Schmidt conjugate direction approach without derivative to other minimization methods is conducted and further analytical and computational details are provided.