Stochastic variational inference for GARCH models

被引：0
|
作者：
Xuan H. [1 ]
Maestrini L. [2 ]
Chen F. [1 ,3 ]
Grazian C. [4 ,5 ]
机构：
[1] School of Mathematics and Statistics, University of New South Wales, Anita B. Lawrence Centre, Kensington, 2052, NSW
[2] Research School of Finance, Actuarial Studies and Statistics, The Australian National University, Building 26C, Kingsley Street, Canberra, 2601, ACT
[3] UNSW Data Science Hub, University of New South Wales, Anita B. Lawrence Centre, Kensington, 2052, NSW
[4] School of Mathematics and Statistics, University of Sydney, Carslaw Building, Camperdown, 2006, NSW
[5] ARC Training Centre in Data Analytics for Resources and Environments, University of Sydney, Biomedical Building, South Eveleigh, 2015, NSW
来源：
Statistics and Computing | 2024年 / 34卷 / 1期
关键词：
Bayesian inference; Financial time series; Skewed t innovation; Stochastic gradient descent; Variational Bayes;
D O I：
10.1007/s11222-023-10356-7
中图分类号：
学科分类号：
摘要：
Stochastic variational inference algorithms are derived for fitting various heteroskedastic time series models. We examine Gaussian, t, and skewed t response GARCH models and fit these using Gaussian variational approximating densities. We implement efficient stochastic gradient ascent procedures based on the use of control variates or the reparameterization trick and demonstrate that the proposed implementations provide a fast and accurate alternative to Markov chain Monte Carlo sampling. Additionally, we present sequential updating versions of our variational algorithms, which are suitable for efficient portfolio construction and dynamic asset allocation. © 2023, The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature.
引用
下载
收藏
相关论文
共 50 条
  • [41] GARCH dependence in extreme value models with Bayesian inference
    Zhao, Xin
    Scarrott, Carl John
    Oxley, Les
    Reale, Marco
    MATHEMATICS AND COMPUTERS IN SIMULATION, 2011, 81 (07) : 1430 - 1440
  • [42] Statistical inference for mixture GARCH models with financial application
    Maddalena Cavicchioli
    Computational Statistics, 2021, 36 : 2615 - 2642
  • [43] Bayesian inference of multivariate-GARCH-BEKK models
    Livingston, G. C. Jr Jr
    Nur, Darfiana
    STATISTICAL PAPERS, 2023, 64 (05) : 1749 - 1774
  • [44] STATISTICAL INFERENCE FOR MULTIVARIATE RESIDUAL COPULA OF GARCH MODELS
    Chan, Ngai-Hang
    Chen, Jian
    Chen, Xiaohong
    Fan, Yanqin
    Peng, Liang
    STATISTICA SINICA, 2009, 19 (01) : 53 - 70
  • [45] Bayesian inference of multivariate-GARCH-BEKK models
    G. C. Livingston
    Darfiana Nur
    Statistical Papers, 2023, 64 : 1749 - 1774
  • [46] DOUBLY STOCHASTIC MODELS WITH THRESHOLD GARCH INNOVATIONS
    Sheraz, Muhammad
    MATHEMATICAL REPORTS, 2014, 16 (03): : 421 - 430
  • [47] Variational inference for multiplicative intensity models
    Lau, John W.
    Cripps, Edward
    Hui, Wendy
    STATISTICS & PROBABILITY LETTERS, 2020, 161
  • [48] An Instability in Variational Inference for Topic Models
    Ghorbani, Behrooz
    Javadi, Hamid
    Montanari, Andrea
    INTERNATIONAL CONFERENCE ON MACHINE LEARNING, VOL 97, 2019, 97
  • [49] Variational Inference for Sparse and Undirected Models
    Ingraham, John
    Marks, Debora
    INTERNATIONAL CONFERENCE ON MACHINE LEARNING, VOL 70, 2017, 70
  • [50] INFERENCE FOR STOCHASTIC NEURONAL MODELS
    HABIB, MK
    THAVANESWARAN, A
    APPLIED MATHEMATICS AND COMPUTATION, 1990, 39 (03) : S183 - S205
12345