Monitorowanie autokorelacyjnego procesu za pomocą łańcuchów Markowa
Monitoring Autocorrelated Processes Using the Markov Chain Monte Carlo Method
Author(s): Grzegorz Kończak, Małgorzata SzerszunowiczSubject(s): Economy
Published by: Wydawnictwo Uniwersytetu Ekonomicznego w Krakowie
Keywords: autocorrelated processes; Markov chain Monte Carlo; computer simulation
Summary/Abstract: The concept of the statistical control chart was developed in 1924 by W. A. Shewhart. The control chart is a graphical display of a quality characteristic such as sample mean, standard deviation or range. The classical control charts are constructed under such assumptions as the form of distribution and independence, and the normality of the distribution is usually assumed. In many situations we may have reason to doubt the validity of the independence assumption – for example, in chemical processes where consecutive measurements on process characteristics are often highly correlated. The paper presents a proposal for a control chart for monitoring auto-correlated processes. The properties of this control chart were analysed in a Monte Carlo study.
Journal: Zeszyty Naukowe Uniwersytetu Ekonomicznego w Krakowie
- Issue Year: 892/2012
- Issue No: 16
- Page Range: 69-78
- Page Count: 10
- Language: Polish