.png)
584 – Recent Advances in the Analysis of Nonignorable Missing Data
SPC Charts for Detecting Shifts in Variance with Autocorrelated Data
Sponsor: Quality and Productivity Section
Keywords: Autocorrelatted Processes, Average Run Length, Exponentially Weighted Mean Square, Nonstationary Processes, Variance Shift
Nien-Fan Zhang
NIST
Adam L. Pintar
NIST
Processes that arise naturally, e.g., from manufacturing or the environment, often exhibit complicated autocorrelation structures. When monitoring such a process for changes in variance, accounting for that autocorrelation structure is critical. While charts for monitoring the variance of processes of independent observations and some specific autocorrelated processes have been proposed in the past, the chart presented in this article can handle any general stationary process. The performance of the proposed chart was examined through simulations for AR(1) processes and demonstrated with an example.