RobustSTL: A Robust Seasonal-Trend Decomposition Algorithm for Long Time Series

Authors: Qingsong Wen, Jingkun Gao, Xiaomin Song, Liang Sun, Huan Xu, Shenghuo Zhu5409-5416

AAAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments on different synthetic and real-world time series datasets demonstrate that our method outperforms existing solutions.
Researcher Affiliation Industry Qingsong Wen, Jingkun Gao, Xiaomin Song, Liang Sun, Huan Xu, Shenghuo Zhu Machine Intelligence Technology, Alibaba Group Bellevue, Washington 98004, USA {qingsong.wen, jingkun.g, xiaomin.song, liang.sun, huan.xu, shenghuo.zhu}@alibaba-inc.com
Pseudocode Yes Algorithm 1 Robust STL Algorithm Summary
Open Source Code No The paper mentions implementing their algorithm in Python and using CVXOPT, but does not provide a link or statement for their own source code.
Open Datasets Yes One is the supermarket and grocery stores turnover from 2000 to 2009 (Alexandrov et al. 2012)
Dataset Splits No The paper mentions that parameters for baseline algorithms are optimized using cross-validation, but does not specify exact split percentages, sample counts, or detailed splitting methodology.
Hardware Specification No No specific hardware details (e.g., exact GPU/CPU models, processor types, or memory amounts) used for running experiments were mentioned.
Software Dependencies No The paper mentions using Python and the CVXOPT library, as well as R packages forecast and stR, but does not provide specific version numbers for these software dependencies.
Experiment Setup Yes For our proposed Robust STL algorithm, we set the regularization coefficients λ1 = 10, λ2 = 0.5 to control the signal smoothness in the trend extraction, and set the neighborhood parameters K = 2, H = 5 in the seasonality extraction to handle the seasonality shift.