HYBRID CODED GENETIC ALGORITHM TOWARDS DETERMINING SMOOTH TRANSITION AUTOREGRESSIVE MODELS
Abstract
In this paper, the specification of the smooth transition autoregressive (STAR) models as an optimization
problem has been handled with genetic algorithms. In this context a hybrid-coded genetic algorithm is
used. The success of the genetic algorithm based approach is evaluated via a benchmark STAR model
determined by conventional method. Better-fitted models than the benchmark model are obtained with the
proposed approach.
Keywords
References
- Balcombe, K. G. (2005). Model Selection Using Information Criteria and Genetic Algorithms. Computational Economics, 25(3), 207-228.
- Baragona, R., Battaglia, F., & Cucina, D. (2004). Fitting piecewise linear threshold autoregressive models by means
Details
Primary Language
English
Subjects
-
Journal Section
Research Article
Publication Date
April 29, 2016
Submission Date
January 16, 2017
Acceptance Date
-
Published in Issue
Year 2016 Number: 24