Visual Time Series Forecasting: An Image-driven Approach
Abstract
In this work, we address time-series forecasting as a computer vision task. We capture input data as an image and train a model to produce the subsequent image. This approach results in predicting distributions as opposed to pointwise values. To assess the robustness and quality of our approach, we examine various datasets and multiple evaluation metrics. Our experiments show that our forecasting tool is effective for cyclic data but somewhat less for irregular data such as stock prices. Importantly, when using image-based evaluation metrics, we find our method to outperform various baselines, including ARIMA, and a numerical variation of our deep learning approach.
Cite
@article{arxiv.2107.01273,
title = {Visual Time Series Forecasting: An Image-driven Approach},
author = {Naftali Cohen and Srijan Sood and Zhen Zeng and Tucker Balch and Manuela Veloso},
journal= {arXiv preprint arXiv:2107.01273},
year = {2021}
}
Comments
This work was intended as a replacement of arXiv:2011.09052 and any subsequent updates will appear there