Nien Fan Zhang 2006 Metrologia 43 195 doi:10.1088/0026-1394/43/3/002
Nien Fan Zhang
Show affiliationsWeighted mean has been used to combine means from several sets of measurements, which are either from different laboratories or based on different measurement methods. It is also often used in key comparisons and other inter-laboratory comparisons. However, the traditional estimator of the variance of the weighted mean underestimates the variance. Two new variance estimators are proposed with smaller biases, smaller mean square errors and much better coverage probabilities for the correspondingly formed intervals. In metrology, the uncertainties based on Type B evaluation which are not based on statistical analysis of data cannot be ignored. In the later part of this paper, the results are extended to the general case with both Type A and Type B uncertainty components being presented.
06.20.Dk Measurement and error theory
02.50.-r Probability theory, stochastic processes, and statistics
Issue 3 (June 2006)
Received 23 August 2005
Published 28 February 2006
Nien Fan Zhang 2006 Metrologia 43 195
Bruno Dumant et al 1999 Distrib. Syst. Engng. 6 3
J Hietarinta et al 1994 J. Phys. A: Math. Gen. 27 3149
Changsong Zhou and Jürgen Kurths 2005 New J. Phys. 7 18
A Ashtekar and A Magnon 1984 Class. Quantum Grav. 1 L39
Takamitsu Okumura et al 2003 J. Opt. A: Pure Appl. Opt. 5 S269
Tomaž Prosen and Marko Žnidarič 2003 New J. Phys. 5 109
Matti A Hiltunen et al 1999 Distrib. Syst. Engng. 6 103
B G Vaucher et al 1995 Metrologia 32 331
J T Lau et al 2002 New J. Phys. 4 98