Spatiotemporal procedures for the statistical surveillance of spatial autoregressive models with heavy tails. Issue 10 (3rd October 2022)

• Record Type: Journal Article
• Authors: Garthoff, Robert; Otto, Philipp
• Is Part Of: Communications in statistics. Volume 51:Issue 10(2022)
• Page Start: 5709
• Page End: 5737
• Publication Date: 2022-10-03
• Subjects: Monte Carlo simulation -- SAR model with t-distributed error term -- Spatiotemporal process control; Mathematical statistics -- Periodicals; Mathematical statistics -- Data processing -- Periodicals; Digital computer simulation -- Periodicals; 519.5
• DOI: 10.1080/03610918.2020.1779294
• Languages: English
• ISSNs: 0361-0918

Abstract

Abstract: The purpose of this article is the statistical surveillance of spatial autoregressive models, where the observed process is monitored over both space and time. The considered spatial model contains disturbances with heavy tails. The control procedures based on exponential smoothing or cumulative sums are constructed using characteristic quantities including the first and the second moments to monitor both means and covariances. Via Monte Carlo simulation, the in-control upper control limits of the control schemes are derived. In a further simulation study, we compare the detection speed of these procedures in the out-of-control situation.