Start with a normal distribution with mean μ and standard deviation σ. You can observe such a thing begin to form with a simple pachinko board: the balls tend to collect in the middle and less so further from the middle.
The standard normal distribution Z is invented to have mean zero and standard deviation 1. You may standardize your normal data x with Z=(x-μ)/σ.
Name:
Anonymous2009-06-08 4:39
Now you may take any number of normal variables x, standardize them to a set of standard variables z, square those z2, and then add those squares. That procedure forms a chi-square distribution χ2. There is a unique chi-square distribution for using 1 z2, for using 2 different z2, for using 3 different z2, and so on.
Name:
Anonymous2009-06-08 4:47
Form the t-distribution by dividing the Z distribution by the square root of the quotient of a χ2 distribution and one less than the number of z2 that had formed it. For instance, if the χ2 distribution was the consequence of 5 summed z2, the division of Z by square root (χ2/4) would produce the t-distribution. It follows that there is a unique t-distribution for every number of underlying z2 inherent in the χ2 distribution.
Name:
Anonymous2009-06-08 4:52
Form the F-distribution (named for Fisher, a genius and a receiver of knighthood) by forming a ratio of two quotients: a χ2 distribution divided by number of underlying z2 inherent and a second χ2 distribution divided by number of underlying z2 inherent. It follows that there is a unique F-distribution for every different pair of numbers of underlying z2 inherent.
Name:
Anonymous2009-06-08 12:54
And how does this resmble a deity?
Name:
Anonymous2009-06-08 15:16
The different gods and goddesses of Hinduism represent various functions of this One Supreme Divinity, and are not separate gods.
Alternatively, distributions are like pokemon. Start with the pikachu