Abstract
The problem of controlled linear calibration with a multivariate response and univariate explanatory variable is considered. Exact expressions (in terms of functions of Poisson variables) are derived for the mean squared error of the inverse estimator. These expressions are compared with ones previously derived for the mean squared error of the classical estimator. Previous comparisons between the two estimators have suggested that neither dominates the other. We prove that this is true, and provide numerical comparisons of the two estimators' risk functions.
Original language | English |
---|---|
Pages (from-to) | 473-488 |
Number of pages | 16 |
Journal | Scandinavian Journal of Statistics |
Volume | 23 |
Issue number | 4 |
State | Published - Dec 1996 |
Keywords
- Controlled calibration
- Multivariate linear regression
- Poisson variables