Time series forecasting: problem of heavy-tailed distributed noise

被引:3
|
作者
Markiewicz, Marta [1 ]
Wylomanska, Agnieszka [2 ]
机构
[1] Data & AI InPost, Wielicka 28, PL-30552 Krakow, Poland
[2] Wroclaw Univ Sci & Technol, Fac Pure & Appl Math, Hugo Steinhaus Ctr, Wybrzeze Wyspianskiego 27, PL-50370 Wroclaw, Poland
关键词
Time series forecasting; SARIMAX; Heavy-tailed distribution; Regression tree; Random forest;
D O I
10.1007/s12572-021-00312-x
中图分类号
T [工业技术];
学科分类号
08 ;
摘要
Time series forecasting has been the area of intensive research for years. Statistical, machine learning or mixed approaches have been proposed to handle this one of the most challenging tasks. However, little research has been devoted to tackle the frequently appearing assumption of normality of given data. In our research, we aim to extend the time series forecasting models for heavy-tailed distribution of noise. In this paper, we focused on normal and Student's t distributed time series. The SARIMAX model (with maximum likelihood approach) is compared with the regression tree-based method-random forest. The research covers not only forecasts but also prediction intervals, which often have hugely informative value as far as practical applications are concerned. Although our study is focused on the selected models, the presented problem is universal and the proposed approach can be discussed in the context of other systems.
引用
收藏
页码:248 / 256
页数:9
相关论文
共 50 条
  • [1] Time series forecasting: problem of heavy-tailed distributed noise
    Marta Markiewicz
    Agnieszka Wyłomańska
    [J]. International Journal of Advances in Engineering Sciences and Applied Mathematics, 2021, 13 : 248 - 256
  • [2] Testing nonlinearity of heavy-tailed time series
    De Gooijer, Jan G.
    [J]. JOURNAL OF APPLIED STATISTICS, 2024, 51 (13) : 2672 - 2689
  • [3] Wavelet variances for heavy-tailed time series
    Fonseca, R. V.
    Mondal, D.
    Zhang, L.
    [J]. ENVIRONMETRICS, 2019, 30 (06)
  • [4] Online support vector quantile regression for the dynamic time series with heavy-tailed noise
    Ye, Yafen
    Shao, Yuanhai
    Li, Chunna
    Hua, Xiangyu
    Guo, Yanru
    [J]. APPLIED SOFT COMPUTING, 2021, 110
  • [5] Inference in Heavy-Tailed Nonstationary Multivariate Time Series
    Barigozzi, Matteo
    Cavaliere, Giuseppe
    Trapani, Lorenzo
    [J]. JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 2024, 119 (545) : 565 - 581
  • [6] LASSO GUARANTEES FOR β-MIXING HEAVY-TAILED TIME SERIES
    Wong, Kam Chung
    Li, Zifan
    Tewari, Ambuj
    [J]. ANNALS OF STATISTICS, 2020, 48 (02): : 1124 - 1142
  • [7] Extremal quantile autoregression for heavy-tailed time series
    He, Fengyang
    Wang, Huixia Judy
    Zhou, Yuejin
    [J]. COMPUTATIONAL STATISTICS & DATA ANALYSIS, 2022, 176
  • [8] Random Matrix Theory for Heavy-Tailed Time Series
    Heiny J.
    [J]. Journal of Mathematical Sciences, 2019, 237 (5) : 652 - 666
  • [9] Distributed Consensus Multi-Distribution Filter for Heavy-Tailed Noise
    Chang, Guan-Nan
    Fu, Wen-Xing
    Cui, Tao
    Song, Ling-Yun
    Dong, Peng
    [J]. JOURNAL OF SENSOR AND ACTUATOR NETWORKS, 2024, 13 (04)
  • [10] A distributed consensus filter for sensor networks with heavy-tailed measurement noise
    Dong, Peng
    Jing, Zhongliang
    Shen, Kai
    Li, Minzhe
    [J]. SCIENCE CHINA-INFORMATION SCIENCES, 2018, 61 (11)