Chi-square test

From HandWiki
Revision as of 20:27, 4 August 2021 by imported>PolicyEnforcerIA (attribution)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)


If N measurements yi are compared to some model or theory predicting values gi, and if the measurements are assumed normally distributed around gi, uncorrelated and with variances

Error creating thumbnail: Unable to save thumbnail to destination

, then the sum

Error creating thumbnail: Unable to save thumbnail to destination

follows the

Error creating thumbnail: Unable to save thumbnail to destination

(chi-square) distribution with N degrees of freedom. The

Error creating thumbnail: Unable to save thumbnail to destination

test compares s with the integral of the

Error creating thumbnail: Unable to save thumbnail to destination

distribution; if the sum above is equal to the quantile

Error creating thumbnail: Unable to save thumbnail to destination

of the

Error creating thumbnail: Unable to save thumbnail to destination

distribution

Error creating thumbnail: Unable to save thumbnail to destination

then the probability of obtaining s or a larger value in the 'null hypothesis' (i.e. the yi are drawn from a distribution described by the

Error creating thumbnail: Unable to save thumbnail to destination

) is given by

Error creating thumbnail: Unable to save thumbnail to destination

. Integral curves for the

Error creating thumbnail: Unable to save thumbnail to destination

distribution exist in computer libraries or are tabulated in the literature. Note that the test may express little about the inherent assumptions; wrong hypotheses or measurements can, but need not cause large

Error creating thumbnail: Unable to save thumbnail to destination

's. The only statement to make about a measured s is the one above: ``

Error creating thumbnail: Unable to save thumbnail to destination

is the probability of finding a

Error creating thumbnail: Unable to save thumbnail to destination

as large as s or larger, in the null hypothesis.

Rudolf K. Bock, Oct 2000