Improving " Jackknife Instrumental Variable Estimation method" using A class of immun algorithm with practical application

Abstract

Most of the robust methods based on the idea of sacrificing one side versus promotion of another, the artificial intelligence mechanisms try to balance weakness and strength to make the best solutions in a random search technique. In this paper, a new idea is introduced to improve the estimators of parameters of linear simultaneous equation models that resulting from the Jackknife Instrumental Variable Estimation method (JIVE) by using a class of immune algorithm which called Clonal Selection Algorithm (CSA) and better estimates are obtained using one of the robust criterion which is called Mean Absolut Percentage Error (MAPE). The success of intelligence algorithm mechanisms has been proven that used to improve the parameters of linear simultaneous equation models according to user criterion and real data of size n=48.