Method of missing data imputation for multivariate time series

被引：3
|
作者：
Li Z. [1 ]
Zhang F. [1 ]
Wang Y. [1 ]
Tao Q. [1 ]
Li C. [1 ]
机构：
[1] Equipment Management and Unmanned Aerial Vehicle Engineering College, Air Force Engineering University, Xi'an
来源：
| 2018年 / Chinese Institute of Electronics卷 / 40期
关键词：
Least squares support vector machine; Missing data imputation; Multivariate time series; Similarity search;
D O I：
10.3969/j.issn.1001-506X.2018.01.32
中图分类号：
学科分类号：
摘要：
Multivariate time series is a common data type. However, due to many kinds of interference factors, missing data are inevitable, which affects data process and analysis. Firstly, for the series containing missing data, similar series that are of the same class with it are searched to form the training set. Secondly, making use of least squares support vector machine, missing data are filled by univariate and multivariate filling methods respectively. Then, according to the difference between the filling results of univariate and multivariate filling methods, a combined method is proposed. Finally, extensive experiments are conducted. The results show that the proposed method can fill missing data precisely, and can be used in the case where the amount of missing data is relatively large. © 2018, Editorial Office of Systems Engineering and Electronics. All right reserved.
引用
收藏
页码：225 / 230
页数：5
相关论文
共 17 条
  • [1] Bank Z., Abonyi J., Correlation based dynamic time warping of multivariate time series, Expert Systems with Applications, 39, 17, pp. 12814-12823, (2012)
  • [2] Han M., Xu M.L., Ren W.J., Research on multivariate chaotic time series prediction using mRSM model, Acta Automatica Sinica, 40, 5, pp. 822-829, (2014)
  • [3] Wu H.S., Zhang F.M., Zhang C., Et al., Matching similar patterns for multivariate time series, Journal of Applied Sciences, 31, 6, pp. 643-649, (2013)
  • [4] Pree H., Herwig B., Gruber T., Et al., On general purpose time series similarity measures and their use as kernel functions in support vector machines, Information Sciences, 281, 4, pp. 478-495, (2014)
  • [5] Li H.L., Guo C.H., Piecewise cloud approximation for time series mining, Knowledge-Based Systems, 24, 4, pp. 492-500, (2011)
  • [6] Ma Q., Gu Y., Li F.F., Et al., Order-sensitive missing value imputation technology for multi-source sensory data, Journal of Software, 27, 9, pp. 2332-2347, (2016)
  • [7] Lu Z., Zhao J., Liu Y., Et al., Missing data imputation based on maximal variance weight information coefficient for gas flow in steel industry, Control Theory & Applications, 32, 5, pp. 646-654, (2015)
  • [8] Lei C.L., Rui Z.Y., Thermal error modeling and forecasting based on multivariate autoregressive model for motorized spindle, Mechanical Science and Technology for Aerospace Engineering, 31, 9, pp. 1526-1529, (2012)
  • [9] Zoucas F.A.M., Belfiore P., An empirical analysis of a neural network model for the time series forecasting of different industrial segments, International Journal of Applied Decision Sciences, 8, 3, pp. 261-283, (2015)
  • [10] Vapnik V., The Nature of Statistical Learning Theory, (1995)
12