The wrapped skew Gaussian process for analyzing spatio-temporal data

被引:8
|
作者
Mastrantonio, Gianluca [1 ]
Gelfand, Alan E. [2 ]
Lasinio, Giovanna Jona [3 ]
机构
[1] Roma Tre Univ, Via Silvio DAmico 77, I-00145 Rome, Italy
[2] Duke Univ, 223-A Old Chem Bldg,Box 90251, Durham, NC 27708 USA
[3] Sapienza Univ Rome, Ple Aldo Moro 5, I-00185 Rome, Italy
关键词
Directional data; Hierarchical model; Kriging; Markov chain Monte Carlo; Space-time data; Wave directions; PROJECTED NORMAL-DISTRIBUTION; HIDDEN MARKOV-MODELS; DIRECTIONAL-DATA; BAYESIAN-ANALYSIS; CIRCULAR DATA; REGRESSION-MODELS; DISTRIBUTIONS; FORECAST; SPACE;
D O I
10.1007/s00477-015-1163-9
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
We consider modeling of angular or directional data viewed as a linear variable wrapped onto a unit circle. In particular, we focus on the spatio-temporal context, motivated by a collection of wave directions obtained as computer model output developed dynamically over a collection of spatial locations. We propose a novel wrapped skew Gaussian process which enriches the class of wrapped Gaussian process. The wrapped skew Gaussian process enables more flexible marginal distributions than the symmetric ones arising under the wrapped Gaussian process and it allows straightforward interpretation of parameters. We clarify that replication through time enables criticism of the wrapped process in favor of the wrapped skew process. We formulate a hierarchical model incorporating this process and show how to introduce appropriate latent variables in order to enable efficient fitting to dynamic spatial directional data. We also show how to implement kriging and forecasting under this model. We provide a simulation example as a proof of concept as well as a real data example. Both examples reveal consequential improvement in predictive performance for the wrapped skew Gaussian specification compared with the earlier wrapped Gaussian version.
引用
收藏
页码:2231 / 2242
页数:12
相关论文
共 50 条
  • [1] The wrapped skew Gaussian process for analyzing spatio-temporal data
    Gianluca Mastrantonio
    Alan E. Gelfand
    Giovanna Jona Lasinio
    Stochastic Environmental Research and Risk Assessment, 2016, 30 : 2231 - 2242
  • [2] STGP: Spatio-temporal Gaussian process models for longitudinal neuroimaging data
    Hyun, Jung Won
    Li, Yimei
    Huang, Chao
    Styner, Martin
    Lin, Weili
    Zhu, Hongtu
    NEUROIMAGE, 2016, 134 : 550 - 562
  • [3] An additive approximate Gaussian process model for large spatio-temporal data
    Ma, Pulong
    Konomi, Bledar A.
    Kang, Emily L.
    ENVIRONMETRICS, 2019, 30 (08)
  • [4] Dynamic Gaussian process regression for spatio-temporal data based on local clustering
    Binglin WANG
    Liang YAN
    Qi RONG
    Jiangtao CHEN
    Pengfei SHEN
    Xiaojun DUAN
    Chinese Journal of Aeronautics, 2024, 37 (12) : 245 - 257
  • [5] Dynamic Gaussian process regression for spatio-temporal data based on local clustering
    Wang, Binglin
    Yan, Liang
    Rong, Qi
    Chen, Jiangtao
    Shen, Pengfei
    Duan, Xiaojun
    CHINESE JOURNAL OF AERONAUTICS, 2024, 37 (12) : 245 - 257
  • [6] Gaussian Process-based Spatio-Temporal Predictor
    Varga, Balazs
    ACTA POLYTECHNICA HUNGARICA, 2022, 19 (05) : 69 - 84
  • [7] RECORDING, ANALYZING, AND DISPLAYING SPATIO-TEMPORAL POSITION DATA
    PEDERSEN, DM
    PERCEPTUAL AND MOTOR SKILLS, 1970, 30 (03) : 978 - &
  • [8] Monotonic Gaussian Process for spatio-temporal disease progression modeling in brain imaging data
    Nader, Clement Abi
    Ayache, Nicholas
    Robert, Philippe
    Lorenzi, Marco
    NEUROIMAGE, 2020, 205
  • [9] Complexity reduction for Gaussian Process Regression in spatio-temporal prediction
    Dinh-Mao Bui
    Thien Huynh-The
    Lee, Sungyoung
    Yoon, YongIk
    2015 INTERNATIONAL CONFERENCE ON ADVANCED TECHNOLOGIES FOR COMMUNICATIONS (ATC), 2015, : 326 - 331
  • [10] Classification of Gaussian spatio-temporal data with stationary separable covariances
    Karaliute, Marta
    Ducinskas, Kestutis
    NONLINEAR ANALYSIS-MODELLING AND CONTROL, 2021, 26 (02): : 363 - 374