论文信息 - Model-Based Stratification in Inventory Cost Estimation

Model-Based Stratification in Inventory Cost Estimation

This article describes a method for developing an efficiently stratified sampling design for a generalized difference estimator, using a superpopulation model. The effectiveness of model-based stratification is compared to several conventional designs, using a data base from a complete audit of an inventory of 8,069 items. In this application, model-based designs reduce the required sample size from 9% to 40%, compared to the conventional designs.

[1] John Neter,et al. On the Behavior of Statistical Estimators When Sampling Accounting Populations , 1977 .

[2] Morris H. Hansen,et al. Sample survey methods and theory , 1955 .

[3] C. T. Isaki,et al. Survey Design under the Regression Superpopulation Model , 1982 .

[4] C. Cassel,et al. Some results on generalized difference estimation and generalized regression estimation for finite populations , 1976 .

[5] T. Postelnicu,et al. Foundations of inference in survey sampling , 1977 .

[6] Roger L. Wright,et al. Finite Population Sampling with Multivariate Auxiliary Information , 1983 .

[7] R. Kaplan. Statistical Sampling in Auditing with Auxiliary Information Estimators , 1973 .

[8] J. L. Hodges,et al. Minimum Variance Stratification , 1959 .

[9] Carl-Erik Särndal,et al. On π-inverse weighting versus best linear unbiased weighting in probability sampling , 1980 .

[10] Maurice S. Newman. Financial accounting estimates through statistical sampling by computer , 1976 .

[11] A. Harvey. Estimating Regression Models with Multiplicative Heteroscedasticity , 1976 .