SOME EXPERIENCES WITH NEURAL NETWORKS

Citation data:

Conference on Applied Statistics in Agriculture

Publication Year:
1997
Usage 23
Downloads 21
Abstract Views 2
Repository URL:
http://newprairiepress.org/agstatconference/1997/proceedings/15
DOI:
10.4148/2475-7772.1306
Author(s):
Ballou, Lynda L.; Johnson, Dallas E.
Publisher(s):
New Prairie Press
Tags:
Agriculture; Applied Statistics
article description
This paper gives a brief overview of artificial neural networks which may be used to model data similar to the kind where one usually considers regression models. Many practitioners believe that neural networks perform better than regression models for prediction purposes. Some simulations were performed using three different neural net programs, namely Braincel, Ripley's S+ program, and Nychka's S+ program. These simulations reveal some interesting aspects of neural net programs which should be of interest to anyone considering the use of neural net programs to model continuous data.