Model Answer of the Mid-Term Exam of Order

Statistics (MATH 313)

Prof. Magdy E. El-Adll

Department of Mathematics
Faculty of Science
Helwan University
Model Answer of the Mid-Term Exam of MATH 313 Prof. Magdy E. El-Adll

Question I (6 Points)

Let X1:n < X2:n < ... < Xn:n be the order statistics of a random sample of size n from a continuous
population with pdf f (x) and cdf is F (x). Show that:

1. The cdf of the sample maximum, Xn:n , is given by Fn:n (x) = [F (x)]n , −∞ < x < ∞.

2. The cdf of the sample minimum, X1:n , is F1:n (x) = 1 − [1 − F (x)]n , −∞ < x < ∞.

Answer of Question I

1. The cdf of the maximum order statistic, Xn:n , denoted by Fn:n can be derived as
Fn:n (x) = P (Xn:n 6 x)
= P (max{X1 , X2 , ..., Xn } 6 x)
= P (X1 6 x, X2 6 x, ..., Xn 6 x)
Yn
= P (Xi 6 x) by the independence of {X1 , X2 , ..., Xn }
i=1
n
Y
= F (x) by the identical of {X1 , X2 , ..., Xn }
i=1

= [F (x)]n , −∞ < x < ∞. (1)

2. The cdf of the minimum order statistic, X1:n , denoted by F1:n , can be derived in a similar
way as follows:

F1:n (x) = P (X1:n 6 x)

= P (min{X1 , X2 , ..., Xn } 6 x)
= 1 − P (min{X1 , X2 , ..., Xn } > x)
= 1 − P (X1 > x, X2 > x, ..., Xn > x)
Yn
=1− P (Xi > x) by the independence of {X1 , X2 , ..., Xn }
i=1
n
Y
= 1 − (1 − F (x)) by the identical of {X1 , X2 , ..., Xn }
i=1

= 1 − [1 − F (x)]n , −∞ < x < ∞. (2)

1
Model Answer of the Mid-Term Exam of MATH 313 Prof. Magdy E. El-Adll

Question II (4.5 Points)

Let U1:n < U2:n < ... < Un:n be the order statistics of a random sample of size n from the standard
uniform distribution U (0, 1), with cdf F (u) = u, 0 < u 6 1. Answer for the following questions:
(r)
1. The rth −moment of the ith order statistics, µi:n = E[Ui:n
r
], is:

r n!(i + r − 1)! r n! r n! r i!
(a) E[Ui:n ]= (b) E[Ui:n ]= (c) E[Ui:n ]= (d) E[Ui:n ]=
(n + r)!(i − 1)! (n − r)! (n + i)! r!

2. Find the mean and the variance of the sample median, Ũn , when n = 7. That is, compute
E[Ũ7 ] and Var(Ũ7 ).

3. Find the probability that U1:5 exceeds 0.25. (i.e. P (U1:5 > 0.25).

Answer of Question II

1. The rth −moment of the ith order statistics is,

(r) r n!(i + r − 1)!

µi:n = E[Ui:n ]= (3)
(n + r)!(i − 1)!

2. The median is Ũ7 = U4:7 . To get the mean of the median i.e. E[U4:7 ], set r = 1, i = 4,
1
and n = 7 in relation (3), to obtain E[U4:7 ] = . To obtain the variance of Ũ7 = U4:7 , put
2
2
5
r = 2, i = 4, and n = 7 in relation (3) to get E[U4:7 ] = , so that
18
!2
5 1 1
2
V ar (U4:7 ) = E[U4:7 ] − (E[U4:7 ])2 = − = .
18 2 36

3. By relation (2) with F (u) = u, we have

P (U1:5 > 0.25) = 1 − F1:5 (0.25) = (1 − 0.25)5 = 0.2373.

2
Model Answer of the Mid-Term Exam of MATH 313 Prof. Magdy E. El-Adll

Question III (4.5 Points)

Let X1:3 < X2:3 < X3:3 be the order statistics of a random sample of size n = 3 from Pareto
distribution with pdf and cdf , respectively, are
( (
3 x−4 , x > 1; 0, x < 1;
f (x) = and F (x) = −3
0, otherwise. 1 − x , x ≥ 1.

Answer the following:

1. Find, f1:3 (x), the pdf of the sample minimum, X1:3 . What is the distribution of X1:3 ?

2. Find the cdf of the sample median, F2:3 (x) = P (X2:3 ≤ x).

3. Compute the probability that the sample median, X2:3 , of a sample of size n = 3 is at least
the population median m.
Hint: The population median, m, satisfies F (m) = 1 − F (m) = 0.5.

Answer of Question III

1. In general, the pdf of the sample minimum, X1:n , is f1:n (x) = n[1 − F (x)]n−1 f (x). For Pareto
distribution (with ν = 3) and n = 3 we get

2
f1:3 = 3[1 − F (x)]2 f (x) = 3 1 − 1 − x−3 3x−4 = 9x−10 ,

x > 1,

which is the pdf of Pareto distribution with ν = 9.

2. The cdf of the sample median is,

3  
X 3
F2:3 (x) = P (X2:3 ≤ x) = [F (x)]i [1 − F (x)]3−i
i=2
i
3  
X 3
i
3−i
= 1 − x−3 x−3
i=2
i

2 −3
3−2
3 −3
3−3
= 3 1 − x−3 x + 1 − x−3 x
2 3
= 3x−3 1 − x−3 + 1 − x−3

= 1 + 2x−9 − 3x−6 , x ≥ 1. (4)

3
Model Answer of the Mid-Term Exam of MATH 313 Prof. Magdy E. El-Adll

3. The required probability is

3  
X 3
P (X2:3 ≥ m) = 1 − F2:3 (m) = 1 − (F (m))i (1 − F (m))3−i
i=2
i
 3 X3    3
1 3 1 1 1
=1− =1− (4) = 1 − = .
2 i=2
i 2 2 2

1
√
3
Another solution of part 3 can be obtained by noting that F (m) = 2
implies m = 2.
Therefore, by relation (4), we get

√ √ √ 1
P (X2:3 ≤ m) = F2:3 ( 2) = 1 + 2( 2)−9 − 3( 2)−6 = ,
3 3 3

2
1
That is, P (X2:3 ≥ m) = 1 − 2
= 12 .

End of the model answer.