Presentation 2003/7/16
Modeling and Forecasting of Correlated Time Series State Space Models
Yuya Suzuki, Tohta Suko, Toshiyasu Matsushima,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Gaussian and Liner state space model is one of time series analyses. Kalman Filtering is the method to obtain the Bayes Inference on valiances known Gaussian and Liner state space model. Monte Carlo Filtering is the method to calculate the estimated value on general state space model. When we execute such analysis, there sometimes exist plural time series at the same time. In such case, we should calculate the estimated value from these correlated series. In this paper, we propose the expanded state space model, which includes these time series. Next, we propose how to calculate the estimated value by using information of all time series. Morever, to make sure that our proposition is effective; we prove that the expectation of our predicted risk is smaller than that of single time series. We also show some experimental results by simulation.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Time Series / State Space Model / Kalman Filtering / Monte Carlo Filtering / Bayes inference
Paper # IT2003-38
Date of Issue

Conference Information
Committee IT
Conference Date 2003/7/16(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Information Theory (IT)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Modeling and Forecasting of Correlated Time Series State Space Models
Sub Title (in English)
Keyword(1) Time Series
Keyword(2) State Space Model
Keyword(3) Kalman Filtering
Keyword(4) Monte Carlo Filtering
Keyword(5) Bayes inference
1st Author's Name Yuya Suzuki
1st Author's Affiliation Dep. Of Industrial and Management Systems Engineering()
2nd Author's Name Tohta Suko
2nd Author's Affiliation Dep. Of Industrial and Management Systems Engineering
3rd Author's Name Toshiyasu Matsushima
3rd Author's Affiliation Dep. Of Industrial and Management Systems Engineering
Date 2003/7/16
Paper # IT2003-38
Volume (vol) vol.103
Number (no) 215
Page pp.pp.-
#Pages 6
Date of Issue