基于R-Vine Copula的多维混合型数据控制图设计

doi:10.3969/j.issn.1007-7375.2019.05.016

摘要/Abstract

摘要： 多维混合型数据监测问题一直是质量控制和质量管理中的重点和难点。混合型数据包括名义型、顺序型和数值型3种类型。传统的多变量控制图往往只考虑数值型的数据，在应用中存在一定的局限性。同时，在实际场景中，各类变量之间往往存在一定的相关性，这也是在传统控制图中容易被忽略的关键点。本文通过引入Copula-Vine模型，充分利用了顺序型变量的秩相关性，建立了一种新的基于R-Vine Copula的混合型数据控制图（R-Vine Copula control chart， RVC）。通过算例比较，验证了该控制图相对于现有模型在混合型数据监测方面更强的灵活性和有效性。

关键词: 多维混合型数据, 顺序型变量, R-Vine Copula模型, 统计过程控制

Abstract: Multivariable mixed-type data monitoring is a key and difficult point in quality control and quality management. Mixed-type data includes three types:nominal, ordinal and numerical. Traditional multivariable control charts only consider numerical variables and have limitations in applications. At the same time, in real cases, there exists correlations between different variables, which is also a significant factor and easily ignored in traditional control charts. By introducing the Copula-Vine model, the rank of ordinal variables is full used, establishing a new mixed-type data control chart based on R-Vine Copula (R-Vine Copula control chart, RVC). The proposed RVC control chart is then applied to real data to demonstrate its flexibility and effectiveness higher than existing models.

Key words: multivariable mixed-type data, ordinal variables, R-Vine Copula model, statistical processes control

中图分类号:

TP277

张乔微, 李艳婷. 基于R-Vine Copula的多维混合型数据控制图设计[J]. 工业工程, 2019, 22(5): 126-132,149.

ZHANG Qiaowei, LI Yanting. An R-Vine Copula Based Control Chart for Monitoring Multivariable and Mixed-type Data[J]. Industrial Engineering Journal , 2019, 22(5): 126-132,149.

参考文献

[1] HOTELLING H. Multivariate quality control by air testing of sample bombsights//Eisenhart C, Hastay M W, Wallis W A. Techniques of statistical analysis[M]. New York:McGraw,1947.
[2] WOODALL W H, NCube M M. Multivariate CUSUM quality-control procedures[J]. Technometrics, 1985, 27(3):285-292
[3] LOWRY C, WOODALL W, CHAMP C, et al. A multivariate exponentially weighted moving average control chart[J]. Technometrics, 1992, 34(1):8
[4] BAKIR S. Distribution-free quality control charts based on signed-rank-like statistics[J]. Communications in Statistics, 2006, 35(4):15
[5] GRAHAM M A, CHAKRABORTI S, HUMAN S W. A nonparametric EWMA sign chart for location based on individual measurements[J]. Quality Engineering, 2011, 23(3):227-241
[6] QIU P. Distribution-free multivariate process control based on log-linear modeling[J]. IIE Transactions, 2008, 40(7):664-677
[7] ZOU C, WANG Z, TSUNG F. A spatial rank-based multivariate EWMA control chart[J]. Naval Research Logistics (NRL), 2012, 59(2):91-110
[8] NING X, TSUNG F. A density-based statistical process control scheme for high-dimensional and mixed-type observations[J]. IIE Transactions, 2012, 44(4):301-311
[9] BREUNIG M M, KRIEGEL H P, NG R T, et al. LOF:identifying density-based local outliers[C/OL]. (2000-05-16). https://www.researchgate.net/publication/221214719_LOF_Identifying_Density-Based_Local_Outliers.
[10] LIU R Y. Control charts for multivariate processes[J]. Publications of the American Statistical Association, 1995, 90(432):1380-1387
[11] DING D, TSUNG F, LI J. Rank-based process control for mixed-type data[J]. IIE Transactions, 2016:673-683
[12] SKLAR M. Fonctions de repartition à n dimensions et leurs marges[J]. Publication de 1'Institut de Statistique de 1'Universite de Paris, 1959, 8:229-231
[13] SONG P X, LI M, YUAN Y. Joint regression analysis of correlated data using Gaussian Copulas[J]. Biometrics, 2009, 65(1):60-68
[14] FATAHI Amir Afshin, NOOROSSANA Rassoul, DOKOUHAKI Pershang, et al. Copula-based bivariate ZIP control chart for monitoring rare events[J]. Communications in Statistics, 2012, 41(15):18
[15] SUKPARUNGSEE S, KUVATTANA S, BUSABABODHIN P, et al. Bivariate Copulas on the Hotelling's T² control chart[C/OL]. (2017-12-18). https://www.tandfonline.com/doi/abs/10.1080/03610918.2016.1228958.
[16] TRIVEDI P K, ZIMMER D M, TRIVEDI P K, et al. Copula modeling:an introduction for practitioners//Foundations and Trends^® in Econometrics[M].Boston:Now Publishers, 2005.
[17] VERDIER G. Application of Copulas to multivariate control charts[J]. Journal of Statistical Planning & Inference, 2013, 143(12):2151-2159
[18] LIU R Y, SINGH K, TENG J H. DDMA-charts:nonparametric multivariate moving average control charts based on data depth[J]. Allgemeines Statistisches Archiv, 2004, 88(2):235-258
[19] KOSMIDIS I, KARLIS D. Model-based clustering using Copulas with applications[J]. Statistics & Computing, 2016, 26(5):1079-1099
[20] MIN A, CZADO C. Bayesian inference for multivariate Copulas using Pair-Copula constructions[J]. Journal of Financial Econometrics, 2010, 8(4):511-546
[21] BRECHMANN E C, SCHEPSMEIER U. Modeling dependence with C-and D-vine Copulas:the R package CDVine[J]. Journal of Statistical Software, 2013, 52(3):1-27
[22] SCHMIDL D, CZADO C, HUG S, et al. A Vine-Copula based adaptive MCMC sampler for efficient inference for dynamic systems[J]. Bayesian Analysis, 2013, 8(1):1-22
[23] BEDFORD T, COOKE R M. Vines:a new graphical model for dependent random variables[J]. Annals of Statistics, 2002, 30(4):1031-1068
[24] AAS K, CZADO C, FRIGESSI A, et al. Pair-Copula constructions of multiple dependence[J]. Insurance:Mathematics and Economics, 2009, 44(2):182-198
[25] ZILKO A A, KUROWICKA D. Copula in a multivariate mixed discrete-continuous model[J]. Computational Statistics & Data Analysis, 2016, 103:28-55
[26] KIM D, KIM J M, LIAO S M, et al. Mixture of D-Vine Copulas for modeling dependence[J]. Computational Statistics & Data Analysis, 2013, 64(4):1-19
[27] 郑文静, 李绍军, 蒋达. D-vine copulas混合模型及其在故障检测中的应用[J]. 化工学报, 2017, 68(7):2851-2858 ZHENG Wenjing, LI Shaojun, JIANG Da. Mixture of D-vine copulas model and its application in fault detection[J]. CIESC Journal, 2017, 68(7):2851-2858
[28] DISSMANN J, BRECHMANN E C, CZADO C, et al. Selecting and estimating regular vine Copula and application to financial returns[J]. Computational Statistics & Data Analysis, 2013, 59:52-69
[29] FINK H, KLIMOVA Y, CZADO C, et al. Regime switching vine Copula models for global equity and volatility indices[J]. Econometrics, 2017, 5(1):3
[30] JOE H. Families of m-variate distributions with given margins and m (m-1)/2 bivariate dependence parameters[J]. Lecture Notes-Monograph Series, 1996:120-141
[31] AKAIKE H. Information theory and an extension of the maximum likelihood principle//Springer series in statistics book series[M]. New York:Springer, 1973.
[32] Bozdogan H. Model selection and Akaike's information criterion (AIC):the general theory and its analytical extensions[J]. Psychometrika, 1987, 52(3):345-370
[33] Schwarz G. Estimating the dimension of a model[J]. The annals of statistics, 1978, 6(2):461-464
[34] LIEBSCHER E. Semiparametric density estimators using Copulas[J]. Communication in Statistics-Theory and Methods, 2005, 34(1):59-71
[35] EPANECHNIKOV V A. Non-parametric estimation of a multivariate probability density[J]. Theory of Probability & Its Applications, 1969, 14(1):153-158
[36] SILVERMAN B W. Density estimation for statistics and data analysis[M]. UK:Routledge, 2018.
[37] DUA, D, GRAFF, C. UCI machine learning repository[DB/OL].[2018-11-01]. http://archive.ics.uci.edu/ml.
[38] FAWCETT T. An introduction to ROC analysis[J]. Pattern recognition letters, 2006, 27(8):861-874