Advances and Applications in Statistics
Volume 52, Issue 2, Pages 97 - 119
(February 2018) http://dx.doi.org/10.17654/AS052020097 |
|
DMA CHART MONITORING OF THE FIRST INTEGER-VALUED AUTOREGRESSIVE PROCESSES OF POISSON COUNTS
Suganya Phantu, Saowanit Sukparungsee and Yupaporn Areepong
|
Abstract: In this paper, explicit formulas are proposed to evaluate the performance characteristics of the double moving average control chart (DMA chart) for the first order integer-valued autoregressive (INAR(1)) model and to check the accuracy of explicit formulas by means of Monte Carlo (MC) simulation. The characteristics of control charts are frequently measured by average run length (ARL), which is the expectation of the samples taken before a system signals that it is out of control. The results obtained from the proposed explicit formulas of the average run length (ARL) are in excellent agreement with the results obtained from the MC. The performance of the DMA chart gets better as the value of the span (w) decreases for upward shifts. The numerical results showed that when the process is out of control with increasing shifts, the DMA chart performs better as the value of the span (w) decreases. |
Keywords and phrases: average run length, average delay time, DMA chart, INAR(1) model. |
|
Number of Downloads: 459 | Number of Views: 3471 |
|