Overfitting control inside cascade correlation neural networks applied to electricity contract price prediction
DOI:
https://doi.org/10.22395/rium.v14n26a10Keywords:
time series forecast, cascade correlation, neural networks, electricity market of ColombiaAbstract
Prediction of electricity prices is considered a difficult task due to the number and complexity of factors that influence their performance, and their relationships. Neural networks cascade correlation - CASCOR allows to do a constructive learning and it captures better the characteristics of the data; however, it has a high tendency to overfitting. To control overfitting in some areas regularization techniques are used. However, in the literature there are no studies that: i) use regularization techniques to control overfitting in CASCOR networks, ii) use CASCOR networks in predicting of electrical series iii) compare the performance with traÂditional neural networks or statistical models. The aim of this paper is to model and predict the behavior of the price series of electricity contracts in Colombia, using CASCOR networks and controlling the overfitting by regularization techniques
Downloads
How to Cite
Issue
Section
License
The total or partial reproduction of the contents of the journal for educational, research, or academic purposes is authorized as long as the source is cited. For reproduction for other purposes, express authorization from the Sello Editorial Universidad de MedellÃn is required.