Extension To Dependent Test Statistics
In cases where the tests are not independent, the null distribution of X2 is more complicated. A common strategy is to approximate the null distribution with a scaled χ2-distribution random variable. Different approaches may be used depending on whether or not the covariance between the different p-values is known.
Brown's method is designed to combine dependent p-values with known covariance while Kost's method is designed to combine dependent p-values with unknown covariance.
Read more about this topic: Fisher's Method
Famous quotes containing the words extension, dependent, test and/or statistics:
“The desert is a natural extension of the inner silence of the body. If humanity’s language, technology, and buildings are an extension of its constructive faculties, the desert alone is an extension of its capacity for absence, the ideal schema of humanity’s disappearance.”
—Jean Baudrillard (b. 1929)
“Realism holds that things known may continue to exist unaltered when they are not known, or that things may pass in and out of the cognitive relation without prejudice to their reality, or that the existence of a thing is not correlated with or dependent upon the fact that anybody experiences it, perceives it, conceives it, or is in any way aware of it.”
—William Pepperell Montague (1842–1910)
“Unless the people can choose their leaders and rulers, and can revoke their choice at intervals long enough to test their measures by results, the government will be a tyranny exercised in the interests of whatever classes or castes or mobs or cliques have this choice.”
—George Bernard Shaw (1856–1950)
“We ask for no statistics of the killed,
For nothing political impinges on
This single casualty, or all those gone,
Missing or healing, sinking or dispersed,
Hundreds of thousands counted, millions lost.”
—Karl Shapiro (b. 1913)