Signal Extraction and Knowledge Discovery Based on Statistical Modeling
Abstract
In the coming post IT era, the problems of signal extraction and knowledge discovery from huge data sets will become very important. For this problem, the use of good model is crucial and thus the statistical modeling will play an important role. In this paper, we show two basic tools for statistical modeling, namely the information criteria for the evaluation of the statistical models and generic state space model which provides us with a very flexible tool for modeling complex and time-varying systems. As examples of these methods we shall show some applications in seismology and macro economics.
Cite
Text
Kitagawa. "Signal Extraction and Knowledge Discovery Based on Statistical Modeling." International Conference on Algorithmic Learning Theory, 2003. doi:10.1007/978-3-540-39624-6_2Markdown
[Kitagawa. "Signal Extraction and Knowledge Discovery Based on Statistical Modeling." International Conference on Algorithmic Learning Theory, 2003.](https://mlanthology.org/alt/2003/kitagawa2003alt-signal/) doi:10.1007/978-3-540-39624-6_2BibTeX
@inproceedings{kitagawa2003alt-signal,
title = {{Signal Extraction and Knowledge Discovery Based on Statistical Modeling}},
author = {Kitagawa, Genshiro},
booktitle = {International Conference on Algorithmic Learning Theory},
year = {2003},
pages = {3-14},
doi = {10.1007/978-3-540-39624-6_2},
url = {https://mlanthology.org/alt/2003/kitagawa2003alt-signal/}
}