Abstract
A numerical procedure for the analysis of a two-dimensional Markov chain that is characterized by a large state-space and a geometric region of absorbing states is presented. The chain occurs in the analysis of multivariate statistical process control charts. Our method allows for an arbitrary positioning of absorbing states, and uses the independence between horizontal and vertical transitions to avoid the computation of a large probability transition matrix.
Original language | English (US) |
---|---|
Pages (from-to) | 75-79 |
Number of pages | 5 |
Journal | Communications in Statistics Part B: Simulation and Computation |
Volume | 25 |
Issue number | 1 |
DOIs | |
State | Published - Jan 1 1996 |
Externally published | Yes |
Keywords
- Average run length
- Fundamental matrix
- Mean time until absorption
ASJC Scopus subject areas
- Statistics and Probability
- Modeling and Simulation