An M-estimator for reduced-rank system identification

详细信息    查看全文

• 作者：Shaojie Chen ; ^a ; ^{pzcsj76@gmail.com} ; Kai Liu^b ; Yuguang Yang^c ; Yuting Xu^a ; Seonjoo Lee^d ; Martin Lindquist^a ; Brian S. Caffo^a ; Joshua T. Vogelstein^e ; ^f
• 关键词：High dimension ; Image processing ; Parameter estimation ; State-space model ; Time series analysis
• 刊名：Pattern Recognition Letters
• 出版年：2017
• 出版时间：15 January 2017
• 年：2017
• 卷：86
• 期：Complete
• 页码：76-81
• 全文大小：653 K
• 卷排序：86

文摘

The proposed model generalizes the traditional linear dynamical system model. An efficient algorithm is designed for parameter estimation. The model works very well on high-dimensional neuro-imaging data. It is also a standard tool for big time-series data analysis in many domains.