Abstract
We consider the steady state output analysis problem for a process that satisfies a functional central limit theorem. We construct asymptotically valid confidence intervals for the steady-state mean using normalized excursions of the simulated process. An alternative approach splits the output at the last zero crossing of the normalized path. The methods use a bounded memory of size determined by the user.
Original language | English (US) |
---|---|
Pages (from-to) | 681-684 |
Number of pages | 4 |
Journal | Proceedings - Winter Simulation Conference |
Volume | 1 |
State | Published - 2004 |
Externally published | Yes |
Event | Proceedings of the 2004 Winter Simulation Conference - Washington, DC, United States Duration: Dec 5 2004 → Dec 8 2004 |
All Science Journal Classification (ASJC) codes
- Software
- Computer Science Applications
- Modeling and Simulation