Simulation Run Length Control in the Presence of an Initial Transient

Article Abstract:

The feasibility of the full automation of single run simulations including initial transient removal, confidence interval generation, and run length control was investigated. The estimation of the steady state mean of an output sequence from a discrete event simulation was studied. Performance measures considered are bias, confidence interval coverage, mean confidence interval width, mean run length, and mean amount of deleted data. If the output sequence has a strong transient, then inclusion of a test for stationarity results in better performance. If no initial transient is present then the test results in a slightly degraded performance. Tables showing the computed measures of different initial transients are included.

Models, Simulation, Discrete Simulation, Stochastic Model, Performance Prediction, Transient State

---

Simple Dependent Pairs of Exponential and Uniform Random Variables

Article Abstract:

The accurate characterization of input processes for simulation can require the representation of dependencies among input values. A random-coeffecient linear function of two independent exponential variables yielding a third expotential variable is used in the construction of simple, dependent pairs of exponential variables. By employing negative exponentiation, the constructions yield simple multiplicative-based models for dependent uniform pairs. The ranges of dependency allowable in the models are assessed by correlation calculations. All models are suitable for simulation and are free of point and live concentrations of probabilities.

Modeling, Data modeling software, Mathematics of Computing

---

Quantile Estimation in Dependent Sequences

Article Abstract:

Point and interval estimates for a single quantile in a fixed length sequence of stationary dependent observations are investigated. A sequence of maxima can be substituted for an entire sequence in order to reduce storage requirements. Three confidence interval methods that can exploit this sample size reduction were analyzed. Tables of computational results are included.

Estimation, Sampling Error, Approximation

Similar abstracts:

• Abstracts: The impact of environmental policy instruments on the timing of adoption of energy-saving technologies. Environmental regulation and the productivity of Japanese manufacturing industries
• Abstracts: Depletion of fossil fuels and the impacts of global warming. Efficient incomplete international climate agreements
• Abstracts: Accumulative pollution, "clean technology," and policy design. Financial structure and the effectiveness of pollution control in an oligopolistic industry
• Abstracts: Storm warning. Coal production and consumption to rise
• Abstracts: Stackelberg-Nash-Cournot Equilibria: Characterizations and Computations. A Class of Stochastic Mathematical Programs With Correlated Scale Parameters in the Objective Right-Hand Side

This website is not affiliated with document authors or copyright owners. This page is provided for informational purposes only. Unintentional errors are possible.