摘要
利用函数型数据刻画产品的某些特性是目前统计质量控制研究中非常流行的方法。文章讨论了带有独立同分布随机项的一般线性函数型模型,基于分位数回归方法在线监控截距和斜率中位数。结合多元指数加权移动平均控制图,给出一种新的函数型数据在线监控控制图,通过模拟研究给出控制图的可控平均运行长度,并验证了在模型系数发生漂移的情况下控制图的有效性。
Characterizing certain properties of products by linear profiles is a very popular method in statistical quality control research. The paper examines a general linear profile model with independent and identically distributed random variables,and monitors the intercept and slope median online based on quantile regression. Multivariate exponentially weighted moving average(MEWMA) control chart is also used to propose new on-line monitoring linear profiles with a control chart. This paper also proposes in-control average run length of the control chart by means of simulation study,and verifies the effectiveness of the control chart in case of shifts in the model coefficient.
引文
[1] ABREVAYA J. The effects of demographics and maternal behavior on the distribution of birth outcomes[J]. Empirical Economics,2001,26(1):247-257.
[2] BASSETT G W,CHEN H L. Portfolio style:return-based attribution using quantile regression[J]. Empirical Economics,2001,26(1):293-305.
[3] MACHADO J A F,MATA J. Counterfactual decomposition of changes in wage distributions using quantile regression[J].Journal of Applied Econometrics,2005,20(4):445-465.
[4] ANGRIST J,CHERNOZHUKOV V,FERNANDEZ-VAL I.Quantile regression under misspecification,with an application to the US wage structure[J]. Econometrica,2006,74(2):539-563.
[5] KOENKER R,BASSETT G. Regression quantiles[J]. Econometrica,1978,46(1):33-50.
[6] ZOU C L,TSUNG F,WANG Z J. Monitoring general linear profiles using multivariate exponentially weighted moving average schemes[J]. Technometrics,2007,49(4):395-408.
[7] KIM K,MAHMOUD M A,WOODALL W H. On the monitoring of linear profiles[J]. Journal of Quality Technology,2003,35(3):317-328.
[8] KOENKER R,BASSETT G. Robust tests for heteroscedasticity based on regression quantiles[J]. Econometrica,1982,50(1):43-61.
[9] BASSETT G,KOENKER R. An empirical quantile function for linear models with iid errors[J]. Journal of the American Statistical Association,1982,77(378):407-415.
[10] BASSETT G,KOENKER R. Asymptotic theory of least absolute error regression[J]. Journal of the American Statistical Association,1978,73(363):618-622.
[11] QI D Q,WANG Z J,ZI X M,et al. Phase II monitoring of generalized linear profiles using weighted likelihood ratio charts[J]. Computers&Industrial Engineering,2016,94(1):178-187.
[12] LI Z H,WANG Z J. An exponentially weighted moving average scheme with variable sampling intervals for monitoring linear profiles[J]. Computers&Industrial Engineering,2010,59(4):630-637.
[13]LI Z H,ZOU C L,GONG Z,et al. The computation of average run length and average time to signal:an overview[J].Journal of Statistical Computation and Simulation,2014,84(8):1779-1802.
[14] QI D Q,LI Z H,ZI X M,et al. Weighted likelihood ratio chart for statistical monitoring of queueing systems[J].Quality Technology&Quantitative Management,2016,14(1):19-30.
[15] MALEKI M R,AMIRI A,CASTAGLIOLA P. An overview on recent profile monitoring papers(2008—2018)based on conceptual classification scheme[J]. Computers&Industrial Engineering,2018,126(5):705-728.