Combined Neural Networks for Time Series Analysis
Abstract
We propose a method for improving the performance of any net(cid:173) work designed to predict the next value of a time series. Vve advo(cid:173) cate analyzing the deviations of the network's predictions from the data in the training set. This can be carried out by a secondary net(cid:173) work trained on the time series of these residuals. The combined system of the two networks is viewed as the new predictor. We demonstrate the simplicity and success of this method, by apply(cid:173) ing it to the sunspots data. The small corrections of the secondary network can be regarded as resulting from a Taylor expansion of a complex network which includes the combined system. \Te find that the complex network is more difficult to train and performs worse than the two-step procedure of the combined system.
Cite
Text
Ginzburg and Horn. "Combined Neural Networks for Time Series Analysis." Neural Information Processing Systems, 1993.Markdown
[Ginzburg and Horn. "Combined Neural Networks for Time Series Analysis." Neural Information Processing Systems, 1993.](https://mlanthology.org/neurips/1993/ginzburg1993neurips-combined/)BibTeX
@inproceedings{ginzburg1993neurips-combined,
title = {{Combined Neural Networks for Time Series Analysis}},
author = {Ginzburg, Iris and Horn, David},
booktitle = {Neural Information Processing Systems},
year = {1993},
pages = {224-231},
url = {https://mlanthology.org/neurips/1993/ginzburg1993neurips-combined/}
}