Orderings of Linear Models

Article Abstract:

A comparison is made between models which are determined by linear expectations and covariance structure. When covariances are known, several orderings of models coincide. When covariances are unknown, linear models may be compared by the best linear unbiased estimators. Necessary but not sufficient conditions are obtained from reduced criteria.

Comparative analysis

---

Optimality of Blue's in a General Linear Model with Incorrect Design Matrix

Article Abstract:

Conditions are derived under which all linear representations of the best linear unbiased estimator (BLUE) are presented for the Gauss-Markoff model. The robustness of a subset of estimable parametric functionals is developed. Algebraic formulas are provided.

---

Quadratic Estimation in Mixed Linear Models with Two Variance Components

Article Abstract:

Invariant quadratic unbiased estimates are extended to provide easily computable expressions. Linear combinations of just two quadratic forms are used. Further simplification is obtained for an unbalanced and incomplete genetic design.

Similar abstracts:

• Abstracts: Determinants of interest rate swap spreads. Optimal consumption and investment strategies with stochastic interest rates
• Abstracts: Gaussian Estimation of Single-Factor Continuous Time Models of The Term Structure of Interest Rates. Corporate Behavior in Adjusting to Capital Structure and Dividend Targets: An Econometric Study
• Abstracts: Estimation of the Variance Between Units of Given Size from Experimental Data. Large Sample Theory of the Langevin Distribution

This website is not affiliated with document authors or copyright owners. This page is provided for informational purposes only. Unintentional errors are possible.