3 The exponential family

3.4 Transformation of the random variable

Assume that pdf/pmf for the random variable Y belongs to the exponential family. Then will the random variable X under some one-to-one transformation Y=T(X) also belong to the exponential family? To assess this, recall that when transforming a random variable, the rate of change between X and Y needs to be accounted for via the Jacobian when deriving the pdf/pmf for X:

fX(x|θ)=fY(y=T(x)|θ)|dydx|

Substituting the exponential family pdf/pmf for fY(y|θ) and the Jacobian, |T(x)|, gives:

fX(x|θ)=q(y=T(x))|T(x)|exp{θT(x)-κ(θ)}=h(x)exp{θT(x)-κ(θ)}.

Here, h(x) does not depend on θ and is in its own right a pdf/pmf as it is a transformation of the pdf/pmf q(y). Furthermore, since the transform T(x) also does not depend on θ, the pdf/pmf fX(x|θ) also belongs to the exponential family. The function T(x) is called the sufficient statistic for the parameter θ.

Definition 3.4.1.

A statistic t=T(X) is a sufficient statistic for the parameter θ if the conditional distribution of the data given the statistic does not depend on the parameter θ, i.e.:

(x|t,θ)=(x|t)

 
Exercise 3.25
The pdf for the random variable XGamma(α,1) is:

fX(x|α)=1Γ(α)xα-1exp{-x}forx>0andα>0.

Show that this belongs to the exponential family. What is the sufficient statistic for the canonical parameter?