Approximating and simulating the stochastic growth model: Parameterized expectations, neural networks, and the genetic algorithm

✍ Scribed by John Duffy; Paul D McNelis

• Book ID: 104293745
• Publisher: Elsevier Science
• Year: 2001
• Tongue: English
• Weight: 226 KB
• Volume: 25
• Category: Article
• ISSN: 0165-1889
• DOI: 10.1016/s0165-1889(99)00077-9

✦ Synopsis

This paper suggests a new approach to solving the one-sector stochastic growth model using the method of parameterized expectations. The approach is to employ a &global' genetic algorithm search for the parameters of the expectation function followed by a &local' gradient-descent optimization method to ensure "ne-tuning of the approximated solution. We use this search procedure in combination with either polynomial or neural network speci"cations for the expectation function. We "nd that our approach yields highly accurate solutions in the case where an exact analytic solution exists as well as in cases where no closed-form solution exists. Our results further suggest that neural network speci"cations for the expectation function may be preferred to the more commonly used polynomial speci"cations.