计算机科学

计算机科学 ›› 2023, Vol. 50 ›› Issue (11A): 221200131-6.doi: 10.11896/jsjkx.221200131

• 大数据&数据科学 • 上一篇    下一篇

基于spike-and-slab先验的贝叶斯时间序列模型

郭晨蕾1, 李东喜2   

  1. 1 太原理工大学数学学院 山西 晋中 030600
    2 太原理工大学大数据学院 太原 030024
  • 发布日期:2023-11-09
  • 通讯作者: 李东喜(dxli0426@126.com)
  • 作者简介:(1044650626@qq.com)

Bayesian Time-series Model Based on spike-and-slab Prior

GUO Chenlei1, LI Dongxi2   

  1. 1 College of Mathematics,Taiyuan University of Technology,Jinzhong,Shanxi 030600,China
    2 College of Data Science,Taiyuan University of Technology,Taiyuan 030024,China
  • Published:2023-11-09
  • About author:GUO Chenlei,born in 1997,postgra-duate.Her main research interests include variable selection and so on.
    LI Dongxi,born in 1982,Ph.D,associate professor,postgraduate supervisor.His main research interests include high dimensional data analysis,data mining,machine learning,biostatistics and biological mathematics.

PDF (PC)

547

摘要： 贝叶斯方法通过引入先验信息并结合似然的方法进行参数估计和变量选择,使模型估计和预测结果更为精确。在贝叶斯框架下考虑时间序列之间的相关性,将偏自相关系数融合先验信息,提出基于spike-and-slab先验的贝叶斯层次时间序列模型(Spike-and-slab Prior with Partial Autocorrelation Coefficients,SS-PAC)。SS-PAC模型采用spike-and-slab先验并结合偏自相关系数,实现时间序列滞后阶数的选择、参数估计和预测。基于模拟数据和真实数据的实证研究表明,该模型相较于以往模型在变量选择和预测结果上表现更优。

关键词: 时间序列预测, spike-and-slab先验, 贝叶斯方法, 偏自相关系数, 变量选择

Abstract: Bayesian method makes the results of estimation and prediction more accurate by introducing prior information and combining with likelihood for parameter estimation and variable selection.ABayesian hierarchical time-series model based on spike-and-slab prior with partial autocorrelation coefficients(SS-PAC ) is proposed under the Bayesian framework,considering the correlation between time series,fusing with the partial autocorrelation coefficient and prior information,the SS-PAC model uses spike-and-slab prior and partial autocorrelation coefficient to realize the selection,parameter estimation and prediction of time series lag order.Empirical research through simulated data and real data shows that the model performs better than previous models in variable selection and prediction results.

Key words: Time-series prediction, Spike-and-slab prior, Bayesian method, Partial autocorrelation coefficient, Variable selection

中图分类号: 

  • O212.8
[1]DE GOOIJER J G,HYNDMAN R J.25 years of time seriesforecasting[J].International Journal of Forecasting,2006,22(3):443-473.
[2]TIBSHIRANI R.Regression shrinkage and selection via the lasso[J].Journal of the Royal Statistical Society:Series B(Methodo-logical),1996,58(1):267-288.
[3]FAN J,LI R.Variable selection via nonconcave penalized likelihood and its oracle properties[J].Journal of the American Statistical Association,2001,96(456):1348-1360.
[4]ZOU H,HASTIE T.Regularization and variable selection via the elastic net[J].Journal of the Royal Statistical Society:Series B(Statistical Methodology),2005,67(2):301-320.
[5]ZOU H.The adaptive lasso and its oracle properties[J].Journal of the American Statistical Association,2006,101(476):1418-1429.
[6]VERBESSELT J,ROBINSON A,STONE C,et al.Forecasting tree mortality using change metrics derived from MODIS satellite data[J].Forest Ecology and Management,2009,258(7):1166-1173.
[7]ZHANG C M,ZHANG Z J.Regularized estimation of hemodynamic response function for fMRI data[J].Statistics and Its Interface,2010,3(1):15-31.
[8]MITCHELL T J,BEAUCHAMP J J.Bayesian variable selection in linear regression[J].Am Stat Assoc,1988,83(404):1023-1032.
[9]GEORGE E I,MCCULLOCH R E.Variable selection via gibbs sampling[J].Am Stat Assoc,1993,88(423):881-889.
[10]KUO L,MALLICK B.Variable selection for regression models[J].Sankhyā Indian J Stat Ser B,1998,66(1):65-81.
[11]YANG A,XIANG J,SHU L,et al.Sparse Bayesian Variable Selection with Correlation Prior for Forecasting Macroeconomic Variable using Highly Correlated Predictors[J].Computational Economics,2017,51(2):323-338.
[12]FRANKE P M,HUNTLEY B,PARNELL A C.Frequency selection in paleoclimate time series:A model-based approach incorporating possible time uncertainty[J].Environmetrics,2018,29(2):1-19.
[13]LI Y,LUND R,HEWAARACHCHI A.Multiple changepoint detection with partial information on changepoint times[J].Electronic Journal of Statistics,2019,13(2):2462-2520.
[14]CHEN C W S,LIU F C,PINGAL A C.Integer-valued transfer function models for counts that show zero inflation[J].Statistics &Probability Letters,2023,193(1):109701.
[15]NELDER J,WEDDERBURN R.Generalized Linear Models[J].Journal of the Royal Statistical Society,1972(1):370-384
[16]SAMORODNITSKY S,HOADLEY K A,LOCK E F.A hierar-chical spike-and-slab model for pan-cancer survival using pan-omic data[J].BMC Bioinformatics,2022,23(1):235.
[17]LIU J S,XIA Q.Bayesian Statistical Method Based on MCMC Algorithm [M].Beijing:Science Press,2016.
[18]JOSHUAS.A Conceptual Introduction to Markov Chain Monte Carlo Methods[J].arXiv:Other Statistics,2019.
[19]POSCH K,ARBEITER M,PILZ J.A novel Bayesian approach forvariable selection in linear regression models[J].Computational Statistics and Data Analysis,2020,144(1):106881.
[20]OUYANG L,PARK C,MA Y,et al.Bayesian hierarchical modelling for process optimization[J].International Journal of Production Research,2020,59(15):4649-4669.
[21]LIN Z,VEERABHADRAN B.Bayesian hierarchical structured variable selectionmethods with application to molecular inversionprobe studies in breast cancer[J].J R Stat Soc Ser C Appl Stat,2014,63(4):595-620.
[22]MITCHELL T J,BEAUCHAMP J J.Bayesian Variable Selection inLinear Regression[J].Journal of the American Statistical Association,1988,83(404):1023-1032.
[23]LIU Z,ZHOU J L,DONG C L.Bayesian estimation of multiva-riate linear regression change point model based on MCMC algorithm [J].Henan Science,2020,38(8):1210-1214.
[24]VERONIKA R,EMMANUEL L,JOLANDA L,et al.Hierarchical Bayesian formulations for selecting variables in regression models[J].Statistics in Medicine,2012,31(11/12):1221-1237.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发