ABSTRACT
In this study, an alternative method has been proposed for the parameter estimation in non-linear regression. This method is the genetic algorithms technique which is widely used in recent years. Unlike other parameter estimation methods, genetic algorithms do not require supplementary information and thus are functional for practical purposes. Genetic algorithms are shown to achieve effective solutions compared to other methods in the S-type growth models. The results have established that genetic algorithms with no need for auxiliary information can be used for parameter estimation in non-linear regression.
Key Words: Nonlinear Regression, S-Type Growth Model, Genetic Algorithms
Primary Language | English |
---|---|
Journal Section | Statistics |
Authors | |
Publication Date | August 6, 2010 |
Published in Issue | Year 2004 Volume: 17 Issue: 2 |