NBS Internal Report 74-587; 30p. September 1974.
Least squares; calibration; statistics; measurement restraints in least squares.
This paper presents a discussion of the techniques of statistics that arise in the analysis of data from the comparative experiments used in the calibration of objects relative to a reference group of standards. In calibration one usually measures differences between nominally equal objects and uses the values assigned to the standards as the linear restraint to bring the system up to full rank. The use of the method of least squares in this situation is discussed with examples taken from the calibration work of the National Bureau of Standards.
This summary of statistical techniques is intended for metrologists and was presented at the NBS seminar on High-Efficiency Methods for Dimensional Calibrations given 10-11 June 1974.