Exponential Family
According to the Pitman–Koopman–Darmois theorem, among families of probability distributions whose domain does not vary with the parameter being estimated, only in exponential families is there a sufficient statistic whose dimension remains bounded as sample size increases. Less tersely, suppose are independent identically distributed random variables whose distribution is known to be in some family of probability distributions. Only if that family is an exponential family is there a (possibly vector-valued) sufficient statistic whose number of scalar components does not increase as the sample size n increases.
This theorem shows that sufficiency (or rather, the existence of a scalar or vector-valued of bounded dimension sufficient statistic) sharply restricts the possible forms of the distribution.
Read more about this topic: Sufficient Statistic
Famous quotes containing the word family:
“The law is equal before all of us; but we are not all equal before the law. Virtually there is one law for the rich and another for the poor, one law for the cunning and another for the simple, one law for the forceful and another for the feeble, one law for the ignorant and another for the learned, one law for the brave and another for the timid, and within family limits one law for the parent and no law at all for the child.”
—George Bernard Shaw (1856–1950)