4
$\begingroup$

I fail to understand when we call a family of distribution is complete and when a statistic is complete. What is the difference between both?, Is there a relation between them? Please provide examples along with your answer.

$\endgroup$
6
$\begingroup$

Suppose $X_1,\ldots,X_n \sim \text{i.i.d. } N(\mu,\sigma^2).$

The family of distributions is $$\left\{ N_n\left(\begin{bmatrix} \mu \\ \vdots \\ \mu \end{bmatrix},\sigma^2 \begin{bmatrix} 1 & 0 & 0 & \cdots & 0 \\ 0 & 1 & 0 & \cdots & 0 \\ 0 & 0 & 1 & \cdots & 0 \\ \vdots & \vdots & \vdots & & \vdots \\ 0 & 0 & 0 & \cdots & 1 \end{bmatrix}\right) \quad : \quad \mu\in \mathbb R \right\}$$ (so $\sigma$ is fixed, so that the only difference between one distribution and another in this family is a different value of $\mu$). This is the family of $n$-dimensional normal distributions in which the expected value is that column of $\mu$s and the matrix of covariances is $\sigma^2$ times the $n\times n$ identity matrix. This family of distributions of $n$-tuples is not complete since it admits nontrivial unbiased estimators of $0$; for example $X_1-X_2$ is such an estimator.

Now suppose $T = T(X_1,\ldots,X_n) = X_1+\cdots+X_n.$ It follows that $T\sim N(n\mu,n\sigma^2).$ The family of distributions of $T$ is $\{ N(n\mu,n\sigma^2) : \mu \in\mathbb R\},$ so again $\sigma$ is fixed, so the difference between two members of this family is a different value of $\mu$. This family is complete since it admits no nontrivial unbiased estimators of $0$. The fact that this family of distributions is complete is also expressed by saying that the statistic $T$ is complete. Any time you define a statistic that is a function of $(X_1,\ldots,X_n),$ having already defined a family of distributions for that $n$-tuple, that definition induces another family of distributions for the statistic you have defined. To say that that statistic is complete merely means that that induced family of distributions is complete.

$\endgroup$
  • $\begingroup$ I was having a note in which there was an example on whether the family of Bernoulli distribution with parameter p is complete or not. In notes the solution started by taking n sample points from Bernoulli distribution and a statistic T was defined as T=X1+X2+...+Xn, then T follows binomial distribution with n trials and parameter p. Then it worked out and checked for the completeness of T and at last it concluded that since T is a complete statistics so family of Bernoulli distribution with parameter p is also complete. I think this conclusion is wrong. Am I right? $\endgroup$ – Singh Oct 26 at 4:29

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.