Causal Discovery and Forecasting in Nonstationary Environments with State-Space Models

被引:0
|
作者
Huang, Biwei [1 ]
Zhang, Kun [1 ]
Gong, Mingming [1 ,2 ]
Glymour, Clark [1 ]
机构
[1] Carnegie Mellon Univ, Dept Philosophy, Pittsburgh, PA 15213 USA
[2] Univ Pittsburgh, Dept Biomed Informat, Pittsburgh, PA USA
基金
美国国家科学基金会; 美国国家卫生研究院;
关键词
INFERENCE; NETWORKS;
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
In many scientific fields, such as economics and neuroscience, we are often faced with nonstationary time series, and concerned with both finding causal relations and forecasting the values of variables of interest, both of which are particularly challenging in such nonstationary environments. In this paper, we study causal discovery and forecasting for nonstationary time series. By exploiting a particular type of state-space model to represent the processes, we show that nonstationarity helps to identify causal structure and that forecasting naturally benefits from learned causal knowledge. Specifically, we allow changes in both causal strengths and noise variances in the nonlinear state-space models, which, interestingly, renders both the causal structure and model parameters identifiable. Given the causal model, we treat forecasting as a problem in Bayesian inference in the causal model, which exploits the time-varying property of the data and adapts to new observations in a principled manner. Experimental results on synthetic and real-world data sets demonstrate the efficacy of the proposed methods.
引用
收藏
页数:10
相关论文
共 50 条
  • [21] ON GIBBS SAMPLING FOR STATE-SPACE MODELS
    CARTER, CK
    KOHN, R
    BIOMETRIKA, 1994, 81 (03) : 541 - 553
  • [22] State-Space Models for Online Post-Covid Electricity Load Forecasting Competition
    De Vilmarest, Joseph
    Goude, Yannig
    IEEE OPEN ACCESS JOURNAL OF POWER AND ENERGY, 2022, 9 : 192 - 201
  • [23] Dynamic factors in state-space models for hourly electricity load signal decomposition and forecasting
    Dordonnat, Virginie
    Koopman, Siem Jan
    Ooms, Marius
    2009 IEEE POWER & ENERGY SOCIETY GENERAL MEETING, VOLS 1-8, 2009, : 3926 - 3934
  • [24] State-space models for optical imaging
    Myers, Kary L.
    Brockwell, Anthony E.
    Eddy, William F.
    STATISTICS IN MEDICINE, 2007, 26 (21) : 3862 - 3874
  • [25] State-Space Models for Control and Identification
    2005, Springer Verlag (308):
  • [26] DERIVATION OF STATE-SPACE MODELS OF CRYSTALLIZERS
    DEWOLF, S
    JAGER, J
    VISSER, B
    KRAMER, HJM
    BOSGRA, OH
    ACS SYMPOSIUM SERIES, 1990, 438 : 144 - 158
  • [27] Bootstrapping Periodic State-Space Models
    Guerbyenne, Hafida
    Hamdi, Faycal
    COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION, 2015, 44 (02) : 374 - 401
  • [28] Probabilistic Recurrent State-Space Models
    Doerr, Andreas
    Daniel, Christian
    Schiegg, Martin
    Nguyen-Tuong, Duy
    Schaal, Stefan
    Toussaint, Marc
    Trimpe, Sebastian
    INTERNATIONAL CONFERENCE ON MACHINE LEARNING, VOL 80, 2018, 80
  • [29] Inequality Constrained State-Space Models
    Qian, Hang
    JOURNAL OF BUSINESS & ECONOMIC STATISTICS, 2019, 37 (02) : 350 - 362
  • [30] DISTURBANCE SMOOTHER FOR STATE-SPACE MODELS
    KOOPMAN, SJ
    BIOMETRIKA, 1993, 80 (01) : 117 - 126