Abstract
Uncertainty assignments in the sequel of least squares adjustments comply with the formalism as used in the context of functional relationships.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Here, \(\bar{\beta}_{k}\) acts as a random variable. For the sake of readability, however, capital letter will not be used in case of Greek symbols.
- 2.
Nevertheless, we might wish to average “ad hoc”, i.e. without resorting to the method of least squares. Let us average the measurements of two masses. Ignoring the option that the associated true values differ grossly, the quotations m 1=1/4 kg and m 2=3/4 kg, each being accurate to about ±1 mg, produce the mean \(\bar{m}=1/2\) kg. However, the associated uncertainty exceeds the assumed uncertainties of the input data by a factor of 250 000. The uncertainty has been blown up since the true values of the masses differed. We are sure, uncertainties should be due to measurement errors and not to deviant true values. This drastically exaggerating example elucidates that the averaging of means implies far-reaching consequences if done inappropriately.
References
Works by the Author
M. Grabe, On the assignment of uncertainties within the method of least squares. Poster Paper, Second International Conference on Precision Measurement and Fundamental Constants, Washington, DC, 8–12 June 1981
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer-Verlag Berlin Heidelberg
About this chapter
Cite this chapter
Grabe, M. (2014). Uncertainties of Least Squares Estimators. In: Measurement Uncertainties in Science and Technology. Springer, Cham. https://doi.org/10.1007/978-3-319-04888-8_11
Download citation
DOI: https://doi.org/10.1007/978-3-319-04888-8_11
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-04887-1
Online ISBN: 978-3-319-04888-8
eBook Packages: Physics and AstronomyPhysics and Astronomy (R0)