Multivariate CUSUM Charts with Correlated Observations

  • Cho, Gyo-Young (Department of Statistics, Kyungpook National University) ;
  • Ahn, Young-Sun (Department of Statistics, Kyungpook National University)
  • Published : 2001.04.30

Abstract

In this article we establish multivariate cumulative sum (CUSUM) control charts based on residual vector with correlated observations. We first find the residual vector and its expectation and variance-covariance matrix and then evaluate the average run length (ARL) of the control charts.

References

  1. Forecast stbility and control charts 1979 ASQC Technical Confidence Transactions Abrham, B.;Kartha, C.P.
  2. EWMA and CUSUM Control Charts in the Presence of Correlation unpublished Ph.D. dissertation VanBrackle, L.N.
  3. Technometrics v.34 A Multivariate Exponentially Weighted Moving Average Control Chart Lowry, C.A.;Woodal, W.H.;Champ, C.W.;Rigdon, S.E.
  4. Journal of Quality Technology v.29 Designing a Multivariate EWMA Control Chart Prabu, S.S.;Runger, G.C.
  5. The Canadian Journal of Chemical Engineering v.69 Statistical Process Control Procedures for Correlated Observations Harris, T.J.;Ross, W.H.
  6. Journal of Business and Economic Statistics v.6 Time-Series Modeling for Statistical Process Control Alwan, L.C.;Roberets, H.V.
  7. Technical Report Charting Correlated Process Data Longnecker, M.T.;Ryan, T.P.
  8. Biometrika v.59 An Approach to the Probability Distribution of CUSUM Run Length Brook, D.;Evans, D.A.
  9. Ph.D. Dissertation Topics in Statistical Process Contro Hui, Y.V.
  10. Time-series Analysis, Forecasting and Control Box, G.E.;Jenkins, G.M.