A change point method for linear profile data

Quality and Reliability Engineering International - Tập 23 Số 2 - Trang 247-268 - 2007
Mahmoud A. Mahmoud1, Peter A. Parker2, William H. Woodall3, Douglas M. Hawkins4
1Cairo University, Cairo, Egypt
2NASA Langley Research Center, Hampton, VA, 23681-2199, U.S.A.
3Virginia Tech, Blacksburg, VA 24061‐0439, U.S.A.
4University of Minnesota, Minneapolis, MN 55455, U.S.A.

Tóm tắt

Abstract

We propose a change point approach based on the segmented regression technique for testing the constancy of the regression parameters in a linear profile data set. Each sample collected over time in the historical data set consists of several bivariate observations for which a simple linear regression model is appropriate. The change point approach is based on the likelihood ratio test for a change in one or more regression parameters. We compare the performance of this method to that of the most effective Phase I linear profile control chart approaches using a simulation study. The advantages of the change point method over the existing methods are greatly improved detection of sustained step changes in the process parameters and improved diagnostic tools to determine the sources of profile variation and the location(s) of the change point(s). Also, we give an approximation for appropriate thresholds for the test statistic. The use of the change point method is demonstrated using a data set from a calibration application at the National Aeronautics and Space Administration (NASA) Langley Research Center. Copyright © 2006 John Wiley & Sons, Ltd.

Từ khóa


Tài liệu tham khảo

10.1080/01621459.1958.10501484

10.1080/01621459.1960.10482067

Brown RL, 1975, Techniques for testing the constancy of regression relationships over time (with discussion), Journal of the Royal Statistical Society B, 37, 149

10.1214/aos/1176344133

10.1080/03610929808832238

10.1198/016214503000233

10.1002/(SICI)1099-095X(199911/12)10:6<711::AID-ENV398>3.0.CO;2-T

Farley JU, 1970, A test for a shifting slope coefficient in a linear model, Journal of the American Statistical Association, 65, 1320, 10.1080/01621459.1970.10481167

10.2307/2346352

10.1080/00401706.1983.10487817

10.1093/biomet/76.3.409

10.1016/0378-3758(91)90043-E

10.2307/2290775

10.1006/jmva.1994.1049

10.2307/2291461

Yakir B, 1999, Detecting a change in regression: First‐order optimality, The Annals of Statistics, 27, 1896

10.2307/2346519

10.1016/S0167-9473(00)00068-2

Gallant AR, 1973, Fitting segmented polynomial regression models whose joint points have to be estimated, Journal of the American Statistical Association, 69, 945

10.2307/1928060

10.1016/S0304-4076(99)00025-1

Emerson J, 2001, Testing for structural change of a time trend regression in panel data: Part I, Journal of Propagations in Probability and Statistics, 1, 57

Emerson J, 2002, Testing for structural change of a time trend regression in panel data: Part II, Journal of Propagations in Probability and Statistics, 2, 207

10.1007/BF02289195

10.1080/01621459.1995.10476611

10.1080/00224065.2004.11980276

10.1007/BF00325603

10.1016/S0021-9673(98)00094-6

10.1198/004017004000000455

10.1080/00224065.2000.11980027

AjmaniV.Using EWMA control charts to monitor linear relationships in semiconductor manufacturing.Joint Statistical Meetings San Francisco CA 2003.

10.1080/00401706.1999.10485932

10.1080/00224065.2002.11980134

WilliamsJD WoodallWH BirchJB.Phase I monitoring of nonlinear profiles.Quality and Productivity Research Conference Yorktown Heights NY 2003.

10.1080/00224065.2003.11980225

Neter J, 1990, Applied Linear Statistical Models

Myers RH, 1990, Classical and Modern Regression with Applications

Vostrikova LJ, 1981, Detecting ‘disorder’ in multidimensional random processes, Soviet Mathematics Doklady, 24, 55

Csörgő M, 1997, Limit Theorems in Change‐Point Analysis

10.1080/00224065.1996.11979677

10.2307/2289240

Kendall M, 1977, Inference and Relationship

10.2307/2684184

10.1007/b98886

Ryan TP, 1997, Modern Regression Methods

BurnDA RyanTAJr.A diagnostic test for lack of fit in regression models.ASA Proceedings of the Statistical Computing Section 1983;286–290.

ParkerPA MortonM DraperNR LineWP.A single‐vector force calibration method featuring the modern design of experiments.Proceedings of the American Institute of Aeronautics and Astronautics 39th Aerospace Sciences Meeting and Exhibit Reno NV 2001.

10.1080/00224065.2003.11980233

10.1080/00224065.2005.11980297