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:
“Where there is reverence there is fear, but there is not reverence everywhere that there is fear, because fear presumably has a wider extension than reverence.”
—Socrates (469–399 B.C.)
“For my part, I have no hesitation in saying that although the American woman never leaves her domestic sphere and is in some respects very dependent within it, nowhere does she enjoy a higher station . . . if anyone asks me what I think the chief cause of the extraordinary prosperity and growing power of this nation, I should answer that it is due to the superiority of their woman.”
—Alexis de Tocqueville (1805–1859)
“Tried by a New England eye, or the more practical wisdom of modern times, they are the oracles of a race already in its dotage; but held up to the sky, which is the only impartial and incorruptible ordeal, they are of a piece with its depth and serenity, and I am assured that they will have a place and significance as long as there is a sky to test them by.”
—Henry David Thoreau (1817–1862)
“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)