Lecture #11 Order Statistics

Objectives of this Lecture:

After reading this unit, you should be able to
• understand the basic concept of order statistics (OS)
• understand about the application of OS

Example 11.1.
Let X 1 , X 2 , and X 3 be iid random variable from a triangular distribution with PDF
f (x )= 2 x ; 0< x<1
0 {otherwise
Calculate the probability that the smallest of these X i ’ s exceeds the median of the distribution.

Solution: Let X 1 , X 2 , and X 3 be iid random variable from a triangular distribution with PDF

f ( x )= 2 x ; 0< x<1
{0 otherwise
Then CDF is
x
2
F ( x)=∫ f (x )dx=x ; 0< x <1
0

The median is
M

∫ f ( x)dx= 12
0
M

∫ 2 x dx= 12 So, M =
1
√2
0

By the definition of the smallest order statistics, the PDF becomes for the above density where n=3
3−1
f 1 :3 ( x)=3[1−F ( x)] f ( x )
= 3[1−x 2 ]2. 2 x
3 5
= 6[ x−2 x + x ] 0< x <1
Now,
1
1
[
P X 1 :3 >
]
=∫ f ( x )dx
√2 1 1:3
√2
16
1
= ∫ [ x−2 x 3+ x 5]=
1 8
√2

Hence, the probability that the smallest of these X i ’ s exceeds the median of the distribution.

Order Statistics | Basic Concept | Lecture #11 Section-B

 Lectured by
STT453: Order Statistics Md. Kaderi Kibria, STAT, HSTU

Example 11.2.
A sample of n observations is taken at random from a population with pdf
−x
f (x )= e ; x≥0
0 {
otherwise
Find the PDF of the smallest observations. What are its mean and variance? What is the CDF?

Solution: Let X 1 : n≤ X 2 : n≤. . .≤ X n : n be the order statistics of a random variables the sample of
size n with PDF
−x
f (x )= e ; x≥0
0 { otherwise
The CDF of the distribution becomes
x
F ( x)=∫ f (t)dt
0
x
=∫ e− x dx=1−e− x ; x ≥0
0

The PDF of smallest order statistics for the above distribution becomes
f 1: n ( x)=n[1−F ( x )]n−1 f ( x)
−x n
= n (e ) ; x≥0
According to the definition of mean, the mean of the smallest order statistics becomes
E ( X 1: n )=∫ f ( x ) dx
x
∞
1
=∫ x ne−nx dx =
0 n
According to the definition of variance
∞
2 2 1
V ( X 1 :n )=E ( X 1:n )− { E ( X 1 : n) } =∫ x 2 ne −nx dx− 2
0 n
2 1 1
= 2− 2= 2
n n n
By the definition of CDF, we have
x
F ( X 1 : n)=∫ f 1 : n (x )dx
0
x
−nx −nx
=∫ ne dx=1−e x≥0
0

which is the CDF of the smallest order statistics of standard exponential distribution.

Example 11.3.
A sample of size 4 observations is taken of random from a population with PDF

Order Statistics | Basic Concept | Lecture #11 Section-B

 Lectured by
STT453: Order Statistics Md. Kaderi Kibria, STAT, HSTU

−x
f (x )= e ; x≥0
0 {
otherwise
Find the probability of X 4: 4 ≥3.

Solution: Let X 1 : 4≤ X 2 : 4≤ X 3: 4 ≤X 4 : 4 be the order statistics of a sample of size 4, having the

PDF

−x
f (x )= e ; x≥0
0 { otherwise
and the CDF becomes
x
F ( x)=∫ f (t)dt
0
x
−x −x
=∫ e dx=1−e ; x ≥0
0

According to the PDF of largest order statistics

f n : n ( x )=n [F (x )]n−1 f ( x)
− x 4−1 − x
= 4 ( 1−e ) e
−x −x 3
= 4 e (1−e ) ; x≥0
We know that
∞ ∞
P ( X 4 : 4≥3 )=∫ f 4 : 4 ( x) dx=∫ 4 e− x (1−e− x )3 dx
3 3
−3 4
=1−(1−e )
This is the required probability of X 4: 4 ≥3.

<><><><><><><><> End <><><><><><><><>

References
1. Balakrishnan, N. and A. C. Cohen (2004), Order Statistics and Inference: Estimation
Methods, Academy Press Inc., London.
2. Arnold, B.C., Balakrishnan, N. and Nagaraja, H.N. (2008), First Course in Order
Statistics, John Wiley and Sons, New York.
3. Anderson,T.W. (2003), An Introduction to Multivariate Statistical Analysis, 5th Ed.,
Wiley, N.Y.
4. Johnson, R. A. and Wichern, D. W (2002), Applied Multivariate Statistical Analysis, 5th
Ed., Prentice Hall, N.Y.
5. Izenman, A. J. (2008), Modern Multivariate Statistical Techniques, Springer, USA.
6. David, H.A. (2003), Order Statistics, 3rd Edition. Wiley Series, New Jersey, Canada.

Order Statistics | Basic Concept | Lecture #11 Section-B