Gaussianity Measures for Detecting the Direction of Causal Time Series
Abstract
We conjecture that the distribution of the time-reversed residuals of a causal linear process is closer to a Gaussian than the distribution of the noise used to generate the process in the forward direction. This property is demonstrated for causal AR(1) processes assuming that all the cumulants of the distribution of the noise are defined. Based on this observation, it is possible to design a decision rule for detecting the direction of time series that can be described as linear processes: The correct direction (forward in time) is the one in which the residuals from a linear fit to the time series are less Gaussian. A series of experiments with simulated and real-world data illustrate the superior results of the proposed rule when compared with other state-of-the-art methods based on independence tests.
Cite
Text
Hernández-Lobato et al. "Gaussianity Measures for Detecting the Direction of Causal Time Series." International Joint Conference on Artificial Intelligence, 2011. doi:10.5591/978-1-57735-516-8/IJCAI11-223Markdown
[Hernández-Lobato et al. "Gaussianity Measures for Detecting the Direction of Causal Time Series." International Joint Conference on Artificial Intelligence, 2011.](https://mlanthology.org/ijcai/2011/hernandezlobato2011ijcai-gaussianity/) doi:10.5591/978-1-57735-516-8/IJCAI11-223BibTeX
@inproceedings{hernandezlobato2011ijcai-gaussianity,
title = {{Gaussianity Measures for Detecting the Direction of Causal Time Series}},
author = {Hernández-Lobato, José Miguel and Morales-Mombiela, Pablo and Suárez, Alberto},
booktitle = {International Joint Conference on Artificial Intelligence},
year = {2011},
pages = {1318-1323},
doi = {10.5591/978-1-57735-516-8/IJCAI11-223},
url = {https://mlanthology.org/ijcai/2011/hernandezlobato2011ijcai-gaussianity/}
}