Abstract

In presence of gross errors (blunders) in the observations, we derive the distributions of some statistics defined after a least-squares (LS) adjustment. Quadratic forms of the LS-residuals and LS-estimates of the blunders are considered along with some their combinations. We demonstrate, in two different ways, that the difference between the quadratic form of the LS-residuals and the quadratic form of the LS-estimates of the blunders has a central chi-square distribution, while each of these two quadratic forms has a non-central chi-square distribution with the same non-centrality parameter. Finally we prove the noncentral Fisher distribution of the ratio between the quadratic form of the LS-estimates of the blunders and the difference of the above-mentioned quadratic forms.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call