Biased Variance Estimators for Statistical Inventory Policies.
Abstract
In practice, scientific inventory decisions frequently are based on statistical estimates of the mean mu and variance (s sup 2) of demand. This report examines the efficacy of modified variance estimates of the form beta(s sup 2), where the scaling factor beta is selected according to the criterion of minimizing expected total operating cost per period. Simulation experiments are run with a number of economic parameter settings for both cases of known and statistically estimated mean, and the value of using a beta value other than unity is assessed.
Document Details
- Document Type
- Technical Report
- Publication Date
- Aug 01, 1976
- Accession Number
- ADA028960
Entities
People
- John G. Klincewicz
Organizations
- Yale University