Document Type
Conference Proceeding
Language
eng
Publication Date
2017
Publisher
International Institute of Forecasters
Source Publication
International Symposium on Forecasting
Abstract
Marquette University’s GasDay Project specializes in short-term load forecasting of natural gas demand. Traditionally, this forecasting is done using artificial neural networks and linear regression. This paper examines the viability of using DNNs as component models in the GasDay ensemble. The ensemble of interest is evaluated using weighted MAPE on 88 natural gas data sets and compared to the current GasDay ensemble. The DNN-enhanced ensemble performs better on this metric than the current ensemble.
Recommended Citation
Merkel, Gregory; Povinelli, Richard J.; and Brown, Ronald H., "Deep Neural Network Regression as a Component of a Forecast Ensemble" (2017). Electrical and Computer Engineering Faculty Research and Publications. 286.
https://epublications.marquette.edu/electric_fac/286
Comments
Published version. Published as part of the proceedings of the International Symposium on Forecasting, 2017. Publisher link. © 2017 The Authors. Used with permission.