Advertisement

Bias of estimator of change point detected by a CUSUM procedure

  • Yanhong Wu
Estimation

Abstract

For independent observations from a standard one-parameter exponential family, the estimator of change point after being detected by a CUSUM procedure is defined as the last zero point of the CUSUM process before the alarm time. By assuming that the change occurs far away from beginning and the control limit is large, an explicit form for the bias of estimator is derived conditioning on the change being detected. By further assuming that the change magnitude and its reference value approach zero at the same order, the local second order expansion of the bias is obtained for numerical evaluation. It is found that, surprisingly, even in the normal distribution case, the bias is non-zero when the change magnitude equals to its reference value, in contrast to the continuous time analog and the fixed sample size case. Numerical results show that the approximations are quite satisfactory.

Key words and phrases

Change-point estimator CUSUM procedure quasi-stationary bias random walk theory strong renewal theorem ladder epoches and ladder heights 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Chang, J. T. (1992). On moments of the first ladder height of random walks with small drift,The Annals of Applied Probability,2, 714–738.MathSciNetGoogle Scholar
  2. Hinkley, D. V. (1971). Inference about the change point from cumulative sum tests,Biometrika,58, 509–523.CrossRefMathSciNetGoogle Scholar
  3. Page, E. S. (1954). Continuous inspection schemes,Biometrika,41, 100–114.CrossRefMathSciNetGoogle Scholar
  4. Pollak, M. and Siegmund, D. (1986). Convergence of quasi-stationary to stationary distribution for stochastically monotone Markov processes,Journal of Applied Probability,23, 215–220.CrossRefMathSciNetGoogle Scholar
  5. Siegmund, D. (1979). Corrected diffusion approximations in certain random walk problems,Advances in Applied Probability,11, 701–719.CrossRefMathSciNetGoogle Scholar
  6. Siegmund, D. (1985).Sequential Analysis: Tests and Confidence Intervals, Springer, New York.MATHGoogle Scholar
  7. Srivastava, M. S. and Wu, Y. (1999). Quasi-stationary biases of change point and change magnitude estimation after sequential CUSUM test,Sequential Analysis,18, 203–216.MathSciNetGoogle Scholar
  8. Stone, C. (1965). On moment generating functions and renewal theory,The Annals of Mathematical Statistics,36, 1298–1301.Google Scholar
  9. Wu, Y. (1999). Second order expansions for the moments of minimum point of an unbalanced two-sided normal random walk,Annals of the Institute of Statistical Mathematics,51, 187–200.CrossRefMathSciNetGoogle Scholar

Copyright information

© The Institute of Statistical Mathematics 2004

Authors and Affiliations

  • Yanhong Wu
    • 1
  1. 1.Department of MathematicsUniversity of the PacificStocktonUSA

Personalised recommendations