Abstracts - faqs.org

Abstracts

Business, general

Search abstracts:
Abstracts » Business, general

Optimal mean-squared-error batch sizes

Article Abstract:

The minimal mean squared error (mse) approach for batch size selection may be used when batch size sets the parameters of the estimator of the sample mean's variance. A study is conducted to investigate the optimal mse batch size of the following four estimator types: nonoverlapping-batch-means, overlapping-batch-means, standardized-time-series-area, and nonoverlapping-batch-means combined with standardize-time-series-area estimators. The study demonstrates that the data process determines the asymptotic sample mean variance only through its marginal variance and the total of the autocorrelations weighted by their absolute lags. By incorporating these results into the variance findings of D. Goldman and M.S. Meketon (1986), explicit asymptotic approximations are derived for mse, optimal batch size, optimal mse and robustness of the four estimator types.

Author: Song, Wheyming Tina, Schmeiser, Bruce W.
Publisher: Institute for Operations Research and the Management Sciences
Publication Name: Management Science
Subject: Business, general
ISSN: 0025-1909
Year: 1995
Methods, Economic lot size, Time-series analysis, Time series analysis

User Contributions:

Comment about this article or add new information about this topic:

CAPTCHA


Strong consistency and other properties of the spectral variance estimator

Article Abstract:

A spectral variance estimator that is strongly consistent justifies the construction of a fixed-width confidence interval. This is useful in steady-state simulation output analysis and is an effective way of deciding how long to sample on a single run of the system. Consistent estimation methods provide advantages for large sample sizes as these methods provide more stable confidence interval levels.

Author: Damerdji, Halim
Publisher: Institute for Operations Research and the Management Sciences
Publication Name: Management Science
Subject: Business, general
ISSN: 0025-1909
Year: 1991
Stochastic processes, Theorems (Mathematics)

User Contributions:

Comment about this article or add new information about this topic:

CAPTCHA


Large-sample results for batch means

Article Abstract:

The techniques for estimating the expected results from a simulation model are presented. The batch-means estimator is a useful tool in forecasting results.

Author: Goldsman, David, Melamed, Benjamin, Chien, Chiahon
Publisher: Institute for Operations Research and the Management Sciences
Publication Name: Management Science
Subject: Business, general
ISSN: 0025-1909
Year: 1997
United States, Stationary processes

User Contributions:

Comment about this article or add new information about this topic:

CAPTCHA


Subjects list: Research, Estimation theory, Simulation methods, Simulation
Similar abstracts:
  • Abstracts: Output improves. How technology may save the coal industry. Access economics
  • Abstracts: Decision analysis by augmented probability simulation. Modifying customer expectations of price decreases for a durable product
  • Abstracts: Digital cameras are better, but still waste a lot of time, money. Making the sale; the allure of on-line commerce, its proponents argue, will eventually prove overwhelming
  • Abstracts: The exchange rate crisis of September 1992 and the pricing of Italian financial futures. Using derivatives in major currencies for cross-hedging currency risks in Asian emerging markets
  • Abstracts: Contributions to marketing theory and practice from Industrial Marketing Management. Scientific contributions to the field from the Journal of Business-to-Business Marketing
This website is not affiliated with document authors or copyright owners. This page is provided for informational purposes only. Unintentional errors are possible.
Some parts © 2025 Advameg, Inc.