An iterative fitting procedure is proposed to adjust two-way table estimates under marginal constraints. Adjustment is inversely proportional to the variances of the cell estimates. The procedure guarantees non-negative estimates and is compared to a standard nonlinear adjustment procedure for sample data. The proposed procedure is easier to implement in forest survey where large tables frequently occur. Forest Sci. 31:366-372.
Document Type: Miscellaneous
Project Leader, Multiresource Inventory Techniques, Rocky Mountain Forest and Range Experiment Station, USDA Forest Service, 240 W. Prospect, Fort Collins, CO 80526