基于模糊区间时间序列数据之相关性
东莞理工学院, 计算机科学与技术学院, 广东东莞 523109
| 摘 要： | 互联网的兴起，不仅仅牵动着生活社会的便利改变，同时也带动了数据科学的变革，在数据型态日益多元与数据整合逐渐复杂等环境的冲击中，相依性模糊区间数据俨然成为新时代的探讨方向之一，同时由此延伸的方法探讨也逐渐成为探究的课题。本研究基于相依性模糊区间数据集的衍生，创新提出相依性模糊区间相关系数的演算法，藉以衡量相依性随机变量的相关性强弱，并且经由蒙地卡罗的模拟分析与实际发生的数值分析等方式，验证创新方法所获得的效果。在研究分析结果中，模糊区间自相关系数演算法所得到的结果不仅可以体现变量之间的关系之外，更能体现变量之间的相关性波动情况，据以了解在变量相关性的条件下，两者间数据的分散情况，如此可以更精准的获得变量之间的相关信息，提高数据隐藏信息的挖掘效能，进而提供产业一个全新的相关性衡量方式，据以作为决策者参考之用。 |
| 关 键 词： | 相依数据; 模糊区间自相关系数; 模糊区间互相关系数; 时间序列 |
| DOI： | 10.57237/j.wjms.2026.02.001 |
View:
Downloads:
收稿日期: 2026-06-03
接受日期: 2026-07-01
在线出版日期: 2026-07-28
Evaluation of the Fuzzy Correlation Coefficient for Interval Time Series Data
School of Computer Science and Technology, Dongguan University of Technology, Dongguan 523109, China
| Abstract: | The development of the Internet has advanced data science while increasing the diversity and complexity of data types and integration processes. In this context, dependent fuzzy interval data has emerged as an important research topic. With the advent of big data, data generation has shifted from univariate observations to multi-valued representations. Accordingly, fuzzy data analysis methods based on multi-valued logic have been increasingly investigated for practical statistical applications. Since the advent of the big data era, the nature of data generation has gradually evolved from single-valued data to multi-valued data. To address the analytical needs of practical statistical applications, fuzzy data analysis methods based on multi-valued logic have been increasingly explored and developed. In this study, two algorithms are proposed for estimating correlation coefficients in interval time series data: the fuzzy interval autocorrelation coefficient and the fuzzy interval cross-correlation coefficient. These methods are developed to evaluate fuzzy correlations between random variables represented by count intervals, where count values are expressed as fuzzy interval data. Simulation experiments and numerical case analyses are conducted to assess the performance and accuracy of the proposed methods. The results indicate that the correlation coefficients obtained from fuzzy interval time series data are consistent with those derived from traditional univariate correlation methods. The proposed algorithms are also shown to effectively verify variations between two variables in fuzzy interval time series data. |
| Keywords: | Dependent Data; Fuzzy Interval Autocorrelation Coefficient; Fuzzy Interval Cross-Correlation Coefficient; Time Series |
参考文献
| [1] | 覃玉冰, 邓春林, 杨柳., 基于皮尔逊相关系数的网络舆情评估指标体系构建研究 [J]. 情报探索, 2018(10): 5. https://link.cnki.net/doi/10.13603/j.cnki.51-1621/z.2019.10.009 |
| [2] | 程娟娟. 高校科研与教学关系实证研究——基于皮尔逊相关系数的分析 [J]. 中国高校科技, 2022(10): 7. https://link.cnki.net/doi/10.16209/j.cnki.cust.2022.10.016 |
| [3] | 畲梓航, 徐嘉桦, 姚志玉, 梁伟典. 基于皮尔逊相关系数的网购大数据分析−以天猫佰润居旗舰店交易记录为例 [J]. 韩山师范学院学报, 2020, 41(3): 7. https://link.cnki.net/doi/10.16775/j.cnki.44-1302/G4.2021.03.012 |
| [4] | Ildar Z. Batyrshin, 2019, Constructing Correlation Coefficients from Similarity and Dissimilarity Functions [J], Acta Polytechnica Hungarica, 16(10), p191-204. https://doi.org/10.12700/aph.16.10.2019.10 |
| [5] | 邓蓉婕, 方兆本. 基于斯皮尔曼相关分析的理财产品收益率分析 [J]. 统计与决策, 2019(16): 164-167. https://doi.org/10.13546/j.cnki.tjyjc.2019.16.036 |
| [6] | 张伟锋, 徐维超. 斯皮尔曼简捷相关系数与基尼伽玛相关系数对脉冲噪声的稳健性分析 [J]. 电子世界, 2020(10): 81-82. https://link.cnki.net/doi/10.19353/j.cnki.dzsj.2020.10.042 |
| [7] | Song, Ha Yoon; Park, Seongjin, An Analysis of Correlation between Personality and Visiting Place using Spearman's Rank Correlation Coefficient [J], KSII Transactions on Internet and Information Systems, 2018, 14(5): 1951-1966. https://doi.org/10.3837/tiis.2018.12.027 |
| [8] | Jost P J, Wagenhofer A. Special Section: Economic Analyses in Business Administration [J]. Schmalenbach Business Review, 2018, 70(1): 1-2. https://doi.org/10.1007/s41464-018-0047-9 |
| [9] | Guotao Ma. and Mohammad Rezania, Mohaddeseh Mousavi Nezhad, 2022, Effects of spatial autocorrelation structure for friction angle on the runout distance in heterogeneous sand collapse [J], Transportation Geotechnics, 33: 100705. https://doi.org/10.1016/j.trgeo.2022.100705 |
| [10] | Yang, C. C., 2016, Correlation Coefficient Evaluation for the Fuzzy Interval Data [J], Journal of Business Research, 69(6), 2138~2144. https://doi.org/10.1016/j.jbusres.2015.12.021 |
| [11] | 杨志清. 计数区间之模糊相关系数研究 [J], 统计与决策, 2020, 36(20): 28-32. https://link.cnki.net/doi/10.13546/j.cnki.tjyjc.2020.20.006 |
| [12] | 王鑫蕊, 窦鑫莹, 杨志清, 施建华. 模糊区间点二列相关系数研究 [J], 高师理科学刊, 2022, 42(11): 8-15. https://link.cnki.net/doi/10.19986/j.cnki.23-1418/n.2022.11.002 |
| [13] | 陶洋, 祝小钧, 杨柳. 基于皮尔逊相关系数和信息熵的多传感器数据融合 [J]. 小型微型计算机系统, 2023, 44(5): 1075-1080. https://link.cnki.net/doi/10.19979/j.cnki.issn.1000-7180.2023.05.015 |
| [14] | 季鹏, 陈芳芳, 徐天奇, 等. 基于Spearman相关系数法与有功分量法的高阻接地故障选线方法研究 [J]. 山东电力技术, 2022, 49(12): 8-13. https://link.cnki.net/doi/10.19457/j.cnki.sddlj.2023.10.008 |
| [15] | 孙慧慧. 基于M估计的具有一致相关线性混合效应模型中相关系数的齐性检验 [J]. 应用概率统计, 2013, 29(5): 469-479. http://www.chinadoi.cn/10.3969/j.issn.1001-4268.2013.05.003 |
