The extended-least-squares treatment of correlated data

Book · Sat Dec 31 00:00:00 EST 1994

OSTI ID:125754

Cohen, E R ^[1]; Tuninsky, V S ^[2]

---

1. Rockwell International Science Center, Thousand Oaks, CA (United States)
2. D.I. Mendeleev Institute of Metrology, St. Petersburg (Russian Federation)

A generalization of the extended-least-squares algorithms for the case of correlated discrepant data is given. The expressions of the linear, unbiased, minimum-variance estimators (LUMVE) derived before are reformulated. A posteriori estimates of the variance taking into account the inconsistency of all of the experimental data, have the same form as for the case of non-correlated data. These estimates extend the previous improvement on the {open_quotes}traditional{close_quotes} Birge-ratio procedures to the case of correlated input data.

• OSTI ID: 125754
• Report Number(s): CONF-940603-; TRN: 95:005769-0122
• Resource Relation: Conference: CPEM `94: conference on precision electromagnetic measurements, Boulder, CO (United States), 27 Jun - 1 Jul 1994; Other Information: PBD: 1994; Related Information: Is Part Of 1994 Conference on precision electromagnetic measurements digest; DeWeese, E.; Bennett, G. [eds.]; PB: 577 p.
• Country of Publication: United States
• Language: English

Related Subjects

42 ENGINEERING NOT INCLUDED IN OTHER CATEGORIES
99 MATHEMATICS
COMPUTERS
INFORMATION SCIENCE
MANAGEMENT
LAW
MISCELLANEOUS
EXPERIMENTAL DATA
STATISTICAL MECHANICS
LEAST SQUARE FIT
DATA ANALYSIS
DATA ACQUISITION
STATISTICAL DATA