Robust Designs and Optimality of Least Squares for Regression Problems

Article Abstract:

Bias minimization is achieved in a linear least squares estimator. The corresponding designs are optimal. When least squares estimators with symmetric design do not lead to an optimal solution, the relative efficiency of optimal least squares is not diminishing and has a uniform lower bound.

---

A Characterization of a Universally Optimal Design within a Class of Block Designs

Article Abstract:

A completely symmetric information matrix results from a class of connected binary designs. Characterization is obtained from arbitrary block sizes and arbitrary replications using a symmetric balanced incomplete block design. Universal optimality is achieved.

---

On the D-Optimality of Group Divisible Designs

Article Abstract:

Formulation is provided for the d-optimality of certain types of proper block designs. Parameters are obtained for connected group divisible designs. Optimization within classes is achieved.

Similar abstracts:

• Abstracts: Reforms start to take hold. Operational risk takes center stage for Treasury. ASPs adapt to a changing market
• Abstracts: Court of first resort. Here's looking at you, KPMG. Constructing criticism
• Abstracts: Customers orientation and usage of financial distribution channels
• Abstracts: The impact of communication effectiveness and service quality on relationship commitment in consumer, professional services
• Abstracts: The influence of ethical climate and ethical conflict on role stress in the sales force. Managing culturally diverse buyer-seller relationships: the role of intercultural disposition and adaptive selling in developing intercultural communication competence

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