CH 4 Order Statistics
CH 4 Order Statistics
CH 4 Order Statistics
Since
independent
it follows that
and identically
distributed
Differentiation according to the chain rule gives us the density function of X(n).
1
Example 2 Example 2
Let Y,Y, be a sequence of independent and identically
distributed U(0,1). Furthermore, let X be discrete with
probability function
Theorem 1.2. Consider a set X,X,,Xn of independent and identically According to Theorem 1.2 the density of the median, i.e. X(2), is given by
distributed (continuous) random variables with density function f(x) and
distribution function F(x). For k=1,2,,n, the density of X(k) is given by
Proof. Both a formal and a heuristic proof can be found in the textbook.
2
The joint distribution of X(1) and X(n) Exercise 2.6
Theorem 2.1. Consider a set X,X,,Xn of independent and identically Exercise 2.6. Suppose n points are chosen uniformly and independently of
distributed (continuous) random variables with density function f(x) and each other on the unit disc. Compute the expected value of the area of the
distribution function F(x). The joint density of the extremes is given by annulus obtained by drawing circles through the extremes.
What to do? Let the distance of a point to the center be Rn, n=1,2,,n.
If R(1)=u and R(n)=v we get the following situation:
where R is the distance to the center of the circle and the angle.
3
Exercise 2.6 The joint distribution of the order statistic
Step 2. Let W=R2. It follows from the (univariate) transformation theorem that Theorem 3.1. Consider a set X,X,,Xn of independent and identically
distributed (continuous) random variables with density function f(x) and
distribution function F(x). The joint density of the order statistic is given by
Step 3. The squared distances to the center can thus be seen as n i.i.d.
U(0,1). Let RW represent the range. It then follows from Theorem 2.2 that
and the statement is proven. The marginal distributions are U(0,x) and U(x,1).
Thommy Perlinger, Probability Theory 16
4
Problem 4.4.6 a Problem 4.4.6 a
To find the density of X(3)+X(4) we introduce the auxiliary variable X(3), i.e. Hence, the marginal distribution of U is given by
that is and so