Please wait a minute...
Front. Inform. Technol. Electron. Eng.  2013, Vol. 14 Issue (2): 107-117    DOI: 10.1631/jzus.C1200316
    
Modeling correlated samples via sparse matrix Gaussian graphical models
Yi-zhou He, Xi Chen, Hao Wang
Lancaster University Management School, Lancaster University, Lancaster, LA1 4YU, UK; Department of Statistics, University of South Carolina, Columbia, South Carolina 29201, USA
Download:   PDF(0KB)
Export: BibTeX | EndNote (RIS)      

Abstract  A new procedure of learning in Gaussian graphical models is proposed under the assumption that samples are possibly dependent. This assumption, which is pragmatically applied in various areas of multivariate analysis ranging from bioinformatics to finance, makes standard Gaussian graphical models (GGMs) unsuitable. We demonstrate that the advantage of modeling dependence among samples is that the true discovery rate and positive predictive value are improved substantially than if standard GGMs are applied and the dependence among samples is ignored. The new method, called matrix-variate Gaussian graphical models (MGGMs), involves simultaneously modeling variable and sample dependencies with the matrix-normal distribution. The computation is carried out using a Markov chain Monte Carlo (MCMC) sampling scheme for graphical model determination and parameter estimation. Simulation studies and two real-world examples in biology and finance further illustrate the benefits of the new models.

Key wordsGaussian graphical models      Hyper-inverse Wishart distributions      Mutual fund evaluation      Network     
Received: 09 November 2012      Published: 31 January 2013
CLC:  O212.8  
Cite this article:

Yi-zhou He, Xi Chen, Hao Wang. Modeling correlated samples via sparse matrix Gaussian graphical models. Front. Inform. Technol. Electron. Eng., 2013, 14(2): 107-117.

URL:

http://www.zjujournals.com/xueshu/fitee/10.1631/jzus.C1200316     OR     http://www.zjujournals.com/xueshu/fitee/Y2013/V14/I2/107


Modeling correlated samples via sparse matrix Gaussian graphical models

A new procedure of learning in Gaussian graphical models is proposed under the assumption that samples are possibly dependent. This assumption, which is pragmatically applied in various areas of multivariate analysis ranging from bioinformatics to finance, makes standard Gaussian graphical models (GGMs) unsuitable. We demonstrate that the advantage of modeling dependence among samples is that the true discovery rate and positive predictive value are improved substantially than if standard GGMs are applied and the dependence among samples is ignored. The new method, called matrix-variate Gaussian graphical models (MGGMs), involves simultaneously modeling variable and sample dependencies with the matrix-normal distribution. The computation is carried out using a Markov chain Monte Carlo (MCMC) sampling scheme for graphical model determination and parameter estimation. Simulation studies and two real-world examples in biology and finance further illustrate the benefits of the new models.

关键词: Gaussian graphical models,  Hyper-inverse Wishart distributions,  Mutual fund evaluation,  Network 
[1] Ling ZHOU, Zhi-zhong TAN , Qing-hua ZHANG. A fractional-order multifunctional n-step honeycomb RLC circuit network[J]. Front. Inform. Technol. Electron. Eng., 2017, 18(8): 1186-1196.
[2] Shih-kung LAI, Jhong-you HUANG. Theoretical foundation of a decision network for urban development[J]. Front. Inform. Technol. Electron. Eng., 2017, 18(8): 1033-1039.
[3] Da-hui GAO , Qing-feng WANG , Yong LEI. Distributed fault-tolerant strategy for electric swing system of hybrid excavators under communication errors[J]. Front. Inform. Technol. Electron. Eng., 2017, 18(7): 941-954.
[4] Zhi-yong FENG, Ze-bing FENG, T. Aaron GULLIVER. Joint user association and resource partition for downlink-uplink decoupling in multi-tier HetNets[J]. Front. Inform. Technol. Electron. Eng., 2017, 18(6): 817-829.
[5] He-hao NIU, Bang-ning ZHANG, Dao-xing GUO, Yu-zhen HUANG, Ming-yue LU. Joint cooperative beamforming and artificial noise design for secure AF relay networks with energy-harvesting eavesdroppers[J]. Front. Inform. Technol. Electron. Eng., 2017, 18(6): 850-862.
[6] Lei-ming ZHANG, Long-hao TANG, Yong LEI. Controller area network node reliability assessment based on observable node information*#[J]. Front. Inform. Technol. Electron. Eng., 2017, 18(5): 615-626.
[7] Yong-ping DU, Chang-qing YAO , Shu-hua HUO, Jing-xuan LIU. A new item-based deep network structure using a restricted Boltzmann machine for collaborative filtering[J]. Front. Inform. Technol. Electron. Eng., 2017, 18(5): 658-666.
[8] Yue-bin LUO, Bao-sheng WANG, Xiao-feng WANG, Bo-feng ZHANG. A keyed-hashing based self-synchronization mechanism for port address hopping communication[J]. Front. Inform. Technol. Electron. Eng., 2017, 18(5): 719-728.
[9] Wen-yan CUI, Xiang-ru MENG , Bin-feng YANG , Huan-huan YANG, Zhi-yuan ZHAO. An efficient lossy link localization approach for wireless sensor networks[J]. Front. Inform. Technol. Electron. Eng., 2017, 18(5): 689-707.
[10] Yu-jun Xiao, Wen-yuan Xu, Zhen-hua Jia, Zhuo-ran Ma, Dong-lian Qi. NIPAD: a non-invasive power-based anomaly detection scheme for programmable logic controllers[J]. Front. Inform. Technol. Electron. Eng., 2017, 18(4): 519-534.
[11] Mei-juan Jia, Hui-qiang Wang, Jun-yu Lin, Guang-sheng Feng, Hai-tao Yu. DGTM: a dynamic grouping based trust model for mobile peer-to-peer networks[J]. Front. Inform. Technol. Electron. Eng., 2017, 18(4): 559-569.
[12] Muhammad Asif Zahoor Raja, Iftikhar Ahmad, Imtiaz Khan, Muhammed Ibrahem Syam, Abdul Majid Wazwaz. Neuro-heuristic computational intelligence for solving nonlinear pantograph systems[J]. Front. Inform. Technol. Electron. Eng., 2017, 18(4): 464-484.
[13] Gaurav Bansod, Narayan Pisharoty, Abhijit Patil. BORON: an ultra-lightweight and low power encryption design for pervasive computing[J]. Front. Inform. Technol. Electron. Eng., 2017, 18(3): 332-345.
[14] Erfan Shaghaghi, Mohammad Reza Jabbarpour, Rafidah Md Noor, Hwasoo Yeo, Jason J. Jung. Adaptive green traffic signal controlling using vehicular communication[J]. Front. Inform. Technol. Electron. Eng., 2017, 18(3): 373-393.
[15] Shuo Wang, Jiao Zhang, Tao Huang, Jiang Liu, Yun-jie Liu, F. Richard Yu. FlowTrace: measuring round-trip time and tracing path in software-defined networking with low communication overhead[J]. Front. Inform. Technol. Electron. Eng., 2017, 18(2): 206-219.