Unit-I

(Measures of central tendency, Moments, Skewness and Kurtosis and Random variables)
CO1: Compute and interpret descriptive statistics.
In this Unit, we learn Problems on

I. Mean
II. Median
III. Mode
IV. Raw moments
V. Moments about origin
VI. Central moments
VII. Skewness and Kurtosis
VIII. Single (Uni) discrete random variable
IX. Single (Uni) continuous random variable
X. Concept of Joint r.vs and covariance
I Problems on Mean
S.No Problem
1 The heights of five runners are 140 cm, 157 cm, 146 cm, 151 cm and
167 cm respectively. Find the mean height per runner.
𝑆𝑢𝑚 𝑜𝑓 ℎ𝑒𝑖𝑔ℎ𝑡𝑠 𝑜𝑓 𝑡ℎ𝑒 𝑟𝑢𝑛𝑛𝑒𝑟𝑠
𝑀𝑒𝑎𝑛 ℎ𝑒𝑖𝑔ℎ𝑡 =
𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑟𝑢𝑛𝑛𝑒𝑟𝑠
(140 + 157 + 146 + 151 + 167)
=
5
= 152.2 cm

2 Calculate mean salary

Salaries(in 8 9 4
‘000)
No. of 5 3 2
employees

x f fx
8 5 40
9 3 27
4 2 8
Totals N=10 75
∑ 𝑓𝑥
Mean= =75/10=7.5
𝑁

3. Calculate average age

CI(age groups) 0-10 10-20 20-30
 F 14 34 24

CI x f fx
0-
10 5 14 70
10-
20 15 34 510
20-
30 25 24 600
Totals 72 1180

∑ 𝑓𝑥
Mean= =1180/72=16.39
𝑁

4 If the mean of 9, 8, 10, x, 12 is 15, find the value of x.

(9 + 8 + 10 + x + 12) (39 + x)
Mean of the given numbers = =
5 5

According to the problem, mean = 15 (given).

Therefore, (39 + x)/5 = 15
⇒ 39 + x = 15 × 5
⇒ 39 + x = 75
⇒ 39 - 39 + x = 75 - 39
⇒ x = 36
Hence, x = 36.
5 The mean of 40 numbers was found to be 38. Later on, it was detected
that a number 56 was misread as 36. Find the correct mean of given
numbers by replacing the correct observation
To find mean, we require sum of all observations and total number of observations.
From the given data,we first get those values and later we correct them to get the correct mean
Calculated mean of 40 numbers = 38.
𝑆𝑢𝑚 𝑜𝑓 𝑜𝑏𝑠𝑒𝑟𝑣𝑎𝑡𝑖𝑜𝑛𝑠 ∑𝑥
𝑀𝑒𝑎𝑛 = =
𝑇𝑜𝑡𝑎𝑙 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑜𝑏𝑠𝑒𝑟𝑣𝑎𝑡𝑖𝑜𝑛𝑠 𝑛
∑𝑥
38=
40

Therefore, calculated sum of these numbers = (38 × 40) = 1520.

Correct sum of these numbers
= [1520 - (wrong item) + (correct item)]
= (1520 - 36 + 56)
= 1540.
Therefore, the correct mean = 1540/40 = 38.5.
6 If the mean of 34 individuals age is 14.76 which are distributed in the
following data, calculate x and y values
C.I 10-12 12-14 14-16 16-18 18-20

f 4 12 x 14 y
To find x, we calculate the following table
Mid
Age
points
(C.I) (x) 𝒇𝒙
f
10-12 44
4 11
12-14 156
12 13
14-16 15x
x 15
16-18 238
14 17
18-20 19y
y 19
N=30+x+y 438+15x+19y

Given that N=34

⟹ 30 + 𝑥 + 𝑦 = 34
⟹ 𝑥 + 𝑦 = 4 … . . (1)
We know that
∑𝑓 𝑥
𝑀𝑒𝑎𝑛 =
𝑁
438+15x+19y
14.76 =
34
 ⟹ 438+15x+19y=738
⟹ 15x+19y=300…….(2)
Solving (1) &(2),we get
𝑥 = 3.04 ≅ 3 𝑎𝑛𝑑 𝑦 ≅ 1
II Problems on Median
1 Find the median of the following set of points in a game:
15, 14, 10, 8, 12, 8, 16
First arrange the point values in an ascending order (or descending order).
ascending order is 8, 8, 10, 12, 14, 15, 16
As n=total number of observations=7(odd)
𝑛+1 𝑡ℎ
Median =( ) 𝑡𝑒𝑟𝑚 =12.
2

2 Find the median of the following data:

15, 14, 10, 3, 4, 12, 8, 16
ascending order is 3, 4, 8, 10, 12, 14, 15, 16
As n=total number of observations=8(even)
As n is even, Median is the average of middle most two values.
𝑛 𝑡ℎ 𝑛+2 𝑡ℎ
( ) 𝑡𝑒𝑟𝑚+( ) 𝑡𝑒𝑟𝑚
i.e., 𝑀𝑒𝑑𝑖𝑎𝑛 = 2 2
2
(4)𝑡ℎ 𝑡𝑒𝑟𝑚+(5)𝑡ℎ 𝑡𝑒𝑟𝑚 10+12
= = =11
2 2

3 Calculate Median for the following data:

X 13 2 34 23 4 6
f 3 6 4 7 2 1

To calculate Median, we calculate the following table:

 X f Cf
2 6 6
4 2 8
6 1 9
13 3 12
23 7 19
34 4 23
23+1 𝑡ℎ
As total number of observations is 23(odd), Median will be ( ) 𝑡𝑒𝑟𝑚 = 12𝑡ℎ 𝑡𝑒𝑟𝑚 = 𝟏𝟑
2

4 Find Median
CI 0-10 20-30 15-20 20-25 25-30 30-35 35-40 40-45 45-50
f 15 10 5 15 20 25 10 15 25

To find Median, we calculate the following table

Class frequency C.f
interval
5-10 15 15
10-15 10 25
15-20 5 30
20-25 15 45
 25-30 20 65𝑐𝑓
30-35 25𝑓 90
35-40 10 100
40-45 15 115
45-50 25 140

N=140

𝑁 140
= = 70,
2 2
𝑁
− 𝑐𝑓
𝑀𝑒𝑑𝑖𝑎𝑛 = 𝑙 + 2 ∗𝑐
𝑓
70−65
= 30 + ∗ 5=31
25

III Problems on Mode

1 Calculate Mode for the following:
12,23,34,12,23,45,34,56,34,34,34,67,12,34,15
As the item ‘34’ has maximum repetitions, Mode for the given data=34
2 Calculate Mode for the following:
C.I 10-20 20-30 30-40 40-50 50-60

f 10 20 30 20 5

To find Mode, we calculate the following table:

C.I f

10-20 10

20-30 20𝑓0

𝑙30-40 30𝑓1

40-50 20𝑓2

50-60 5

𝑓1 −𝑓0
𝑀𝑜𝑑𝑒 = 𝑙 + *c
2𝑓1 −𝑓0 −𝑓2
30−20
= 30 + *10
2∗30−20−20

= 35
 Moments
Moment is the average or mean of deviations raised to the required power. Order of the moment is “r”
The two types are
1. Raw moments (𝜇𝑟′ )
a. moments about an arbitrary point ‘a’, here deviations=d=x-a
b. Moments about origin, here deviations=d=x-0
2. Central moments (𝜇𝑟 ), here deviations=d=x- mean of x=x- 𝑥̅
Moments for a normal variable

Central Moments Raw Moments

𝜇1 = 1𝑠𝑡 𝐶𝑒𝑛𝑡𝑟𝑎𝑙 𝑚𝑜𝑚𝑒𝑛𝑡 𝜇1 ′ = 1𝑠𝑡 𝑟𝑎𝑤 𝑚𝑜𝑚𝑒𝑛𝑡 𝑎𝑏𝑜𝑢𝑡 𝑎 𝑝𝑜𝑖𝑛𝑡 a

1 1
= ∑(𝑥 − 𝑥̅ )1 (individual data) = ∑(𝑥 − 𝑎)1 (individual data)
𝑛 𝑛
1 1
= ∑ 𝑓 ∗ (𝑥 − 𝑥̅ )1 (grouped data) = ∑ 𝑓 ∗ (𝑥 − 𝑎)1 (grouped data)
𝑁 𝑁

Note: 𝜇1 = 0 Note: If a=0, 𝜇1 ′ =Mean

𝜇2 = 2𝑛𝑑 𝐶𝑒𝑛𝑡𝑟𝑎𝑙 𝑚𝑜𝑚𝑒𝑛𝑡=Variance= 𝜎 2 𝜇2 ′ = 2𝑛𝑑 𝑟𝑎𝑤 𝑚𝑜𝑚𝑒𝑛𝑡 𝑎𝑏𝑜𝑢𝑡 𝑎 𝑝𝑜𝑖𝑛𝑡 a

1 1
= ∑(𝑥 − 𝑥̅ )2 (individual data) = ∑(𝑥 − 𝑎)2 (individual data)
𝑛 𝑛
 1 1
= ∑ 𝑓 ∗ (𝑥 − 𝑥̅ )2 (grouped data) = ∑ 𝑓 ∗ (𝑥 − 𝑎)2 (grouped data)
𝑁 𝑁

Standard deviation=√𝑉𝑎𝑟𝑎𝑖𝑛𝑐𝑒=𝜎

𝜇3 = 3𝑟𝑑 𝐶𝑒𝑛𝑡𝑟𝑎𝑙 𝑚𝑜𝑚𝑒𝑛𝑡 𝜇3 ′ = 3𝑟𝑑 𝑟𝑎𝑤 𝑚𝑜𝑚𝑒𝑛𝑡 𝑎𝑏𝑜𝑢𝑡 𝑎 𝑝𝑜𝑖𝑛𝑡 a

1 1
= ∑(𝑥 − 𝑥̅ )3 (individual data) = ∑(𝑥 − 𝑎)3 (individual data)
𝑛 𝑛
1 1
= ∑ 𝑓 ∗ (𝑥 − 𝑥̅ )3 (grouped data) = ∑ 𝑓 ∗ (𝑥 − 𝑎)3 (grouped data)
𝑁 𝑁

𝜇4 = 4𝑡h 𝐶𝑒𝑛𝑡𝑟𝑎𝑙 𝑚𝑜𝑚𝑒𝑛𝑡 𝜇4 ′ = 4𝑡h 𝑟𝑎𝑤 𝑚𝑜𝑚𝑒𝑛𝑡 𝑎𝑏𝑜𝑢𝑡 𝑎 𝑝𝑜𝑖𝑛𝑡 a

1 1
= ∑(𝑥 − 𝑥̅ )4 (individual data) = ∑(𝑥 − 𝑎)4 (individual data)
𝑛 𝑛
1 1
= ∑ 𝑓 ∗ (𝑥 − 𝑥̅ )4 (grouped data) = ∑ 𝑓 ∗ (𝑥 − 𝑎)4 (grouped data)
𝑁 𝑁

( Here we are showing the calculation of 3𝑟𝑑 moment about an arbitrary point ‘6’, 3𝑟𝑑
moment about origin and 3𝑟𝑑 central moment
data 2,18,4 X 2 4 9
C.I 0-10 10-20
f 3 2 1
f 5 8

3𝑟𝑑 Calculate
moment d=x-6
X d=x- d=x-
about an Find 𝑑 3 6 𝑑 3
X f 6 𝑑3 f𝑑 3 CI d=x-
arbitrary Find 2 -4 -64 2 3 -4 -64 -192 X f 6 d^3 fd^3
point ‘6’ mean for 0-
18 12 1728 4 2 -2 -8 -16
𝑑3 10 5 5 -1 -1 -5
4 -2 -8 9 1 3 27 27
total 2736 6 -181
 𝜇3′ ∑ 𝑓𝑑3 10-
= 𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑜𝑓 𝑑 3 𝜇3′ = 𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑜𝑓 𝑑 3 = 20 15 8 9 729 5832
𝑁
∑ 𝑑3 ∑ 𝑓(𝑥 − 6) 3 13 728 5827
= = ∑ 𝑓𝑑3
𝑛 𝑁 𝜇3′ = 𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑜𝑓 𝑑 = 3
∑(𝑥 − 6)3 −181 𝑁
= = ∑ 𝑓(𝑥 − 6)3
𝑛 6
1656 = -30.17 =
𝑁
= 5827
3 =
=552 13
= 448.2308

3𝑟𝑑 Calculate
moment d=x-o
X d=x-0 𝑑3 d=x-
about Find 𝑑 3 CI d=x-
2 2 8 X f 0 d^3 fd^3
origin Find X f 0 d^3 fd^3
18 18 5832 2 3 2 8 24
mean for 0-
4 4 64 4 2 4 64 128
𝑑3 10 5 5 5 125 625
total 5904 9 1 9 729 729
Mean for 10-
𝑑3 = 𝜇3′ 6 881
∑ 𝑓𝑑3 20 15 8 15 3375 27000
∑ 𝑑 3 /𝑛 = 𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑜𝑓 𝑑 3
∑ 𝑑3 𝜇3′ = 𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑜𝑓 𝑑 3 = 13 27625
𝑁
= ∑ 𝑓(𝑥 − 0)3 ∑ 𝑓𝑑3
𝑛 = 𝜇3′ = 𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑜𝑓 𝑑 3 =
∑(𝑥 − 0)3 𝑁 𝑁
= 881 ∑ 𝑓(𝑥 − 0)3
𝑛 = =
5904 6 𝑁
= = 146.83 = 27625/13
3
=1968 =2125
3𝑟𝑑 Calculate
central 1.mean
X d=x- fx d=x-
moment 2.Then 3 CI fx d=x-
𝑥̅ 𝑑 X f 𝑥̅ d^3 fd^3
d=x- 𝑥̅ 𝑥̅
2 -6 -216 2 3 6 -1.83 -6.12849 -5.49 X f d^3 fd^3
3.Find 𝑑 3 0-
18 10 1000 4 2 8 0.17 0.004913 0.34
4.Find 10 5 5 25 -6.15 -232.608 -1163.04
4 -4 -64 9 1 9 5.17 138.1884 5.17
mean for 10-
24 720 6 23 0.02
𝑑3 20 15 8 120 3.85 57.06663 456.533
Mean= 𝑥̅ =24/3=8 Mean= 𝑥̅ =23/6=3.83 13 145 -706.509
 ∑ 𝑓𝑑3 Mean= 𝑥̅ =145/13=11.15
𝜇3′ 𝜇3′ = 𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑜𝑓 𝑑 3 =
𝑁
= 𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑜𝑓 𝑑 3 ∑ 𝑓(𝑥 − 𝑥̅ )3 ∑ 𝑓𝑑3
∑ 𝑑3 = 𝜇3′ = 𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑜𝑓 𝑑 3 =
𝑁 𝑁
= 0.02 ∑ 𝑓(𝑥 − 𝑥̅ )3
𝑛 =
∑(𝑥 − 𝑥̅ )3 =
6 𝑁
= = 0.0033 = −706.509/13
𝑛
720 =-54.35
=
3
=240

1 Find out first four moments about a point’4’ for the following data:
2,8,7,4,4.
To find moments about a point ’4’ , we calculate the following table:

(x-a)
X =(x-4) (x-a)^2 (x-a)^3 (x-a)^4

2 -2 4 -8 16

8 4 16 64 256

7 3 9 27 81

4 0 0 0 0

4 0 0 0 0

totals 5 29 83 353
The first four Raw moments are
 1 5
μ1 ′ = ∑(x − a)1 = = 1
n 5
1 29
μ2 ′ = ∑(x − a)2 = = 5.8
n 5
1 83
μ3 ′ = ∑(x − a)3 = = 16.6
n 5
1 353
μ4 ′ = ∑(x − a)4 = = 70.6
n 5

2 Find out first four moments about origin for the following data:
2,8,7,4,4.
To find first four moments about origin, we calculate the following table:

X (x-a)=(X-0) (x-a)^2=X^2 (x-a)^3=X^3 (x-a)^4=X^4

2 2 4 8 16

4096
8 8 64 512
2401
7 7 49 343
256
4 4 16 64
256
4 4 16 64

7025
totals 25 149 991
 The first four Raw moments about origin are
1 25
μ1 ′ = ∑(x)1 = = 5=mean
n 5
1 149
μ2 ′ = ∑ (x)2 = = 29.8
n 5
1 991
μ3 ′ = ∑(x)3 = = 198.2
n 5
1 7025
μ4 ′ = ∑(x)4 = = 1405
n 5
3 Find out first four central moments for the following data:
2,8,7,4,4.
To find first four central moments, we calculate the following table:
X (x-𝑥̅ ) (x-𝑥̅ )^2 (x-𝑥̅ )^3 (x-𝑥̅ )^4

2 -3 9 -27 81

81
8 3 9 27
16
7 2 4 8
1
4 -1 1 -1
1
4 -1 1 -1
Totals
180
=25 0 24 6
 ∑𝑥 25
Mean = 𝑥̅ = = =5
𝑛 5

The first four central moments are

1 0
μ1 = ∑(x − x̅)1 = = 0
n 5
1 24
μ2 = ∑(x − x̅)2 = = 4.8=variance
n 5
1 6
μ3 = ∑(x − x̅)3 = = 1.2
n 5
1 180
μ4 = ∑(x − x̅)4 = = 36
n 5

4 Find out first four moments about a point’18’ for the following data:
C.I 0-10 10-20 20-30 30-40 40-50

f 12 23 24 32 15

To find moments about a point ’18’, we calculate the following table:

x=Mid
C.I f points (x-18) (X-18)^2 (X-18)^3 (X-18)^4 f*(x-18) f(X-18)^2 f(X-18)^3 f(X-18)^4
0-10 12 5 -13 169 -2197 28561 -156 2028 -26364 342732
10-20 23 15 -3 9 -27 81 -69 207 -621 1863
20-30 24 25 7 49 343 2401 168 1176 8232 57624
30-40 32 35 17 289 4913 83521 544 9248 157216 2672672
40-50 15 45 27 729 19683 531441 405 10935 295245 7971615
106 892 23594 433708 11046506
 1 892
𝜇1 ′ = ∑ 𝑓 ∗ (𝑥 − 𝑎)1 = = 8.42,
𝑁 106
1 23594
𝜇2 ′ = ∑ 𝑓 ∗ (𝑥 − 𝑎)2 = =222.59
𝑁 106

1 433708
μ3 ′ = ∑ 𝑓 ∗ (x − a)3 = = 4091.59
𝑁 106

1 11046506
μ4 ′ = ∑ 𝑓 ∗ (x − a)4 = 104212.32
𝑁 106

5 Find out first four moments about origin for the following data:

To find first four moments about origin, we calculate the following table:

x=Mid
C.I f points (x-0) (X-0)^2 (X-0)^3 (X-0)^4 f*(x-0) f(X-0)^2 f(X-0)^3 f(X-0)^4
0-10 12 5 5 25 125 625 60 300 1500 7500
10-20 23 15 15 225 3375 50625 345 5175 77625 1164375
20-30 24 25 25 625 15625 390625 600 15000 375000 9375000
30-40 32 35 35 1225 42875 1500625 1120 39200 1372000 48020000
40-50 15 45 45 2025 91125 4100625 675 30375 1366875 61509375
106 2800 90050 3193000 120076250

1 2800 1 90050
𝜇1 ′ = ∑ 𝑓 ∗ 𝑥 1 = = 26.42, 𝜇2 ′ = ∑ 𝑓 ∗ 𝑥 2 = = 849.53
𝑁 106 𝑁 106
 1 3193000 1 120076250
μ3 ′ = ∑ 𝑓 ∗ 𝑥 3 = = 30122.64, μ4 ′ = ∑ 𝑓 ∗ 𝑥 4 = = 1132794.811
𝑁 106 𝑁 106

6 Find out first four Central moments for the following data:

To find first four Central moments, we calculate the following table:

x=Mid
C.I f points fx (x-𝑥̅ ) (X-𝑥̅ )^2 (X-𝑥̅ )^3 (X-𝑥̅ )^4 f*(x-𝑥̅ ) f(X-𝑥̅ )^2 f(X-𝑥̅ )^3 f(X-𝑥̅ )^4
-
0-10 12 5 60 -21.42 458.8164 210512.4889 -257.04 5505.7968 -117934.1675 2526150
9827.847288
-
10-20 23 15 345 -11.42 130.4164 17008.43739 -262.66 2999.5772 -34255.17162 391194.1
1489.355288
20-30 24 25 600 -1.42 2.0164 -2.863288 4.06586896 -34.08 48.3936 -68.718912 97.58086
30-40 32 35 1120 8.58 73.6164 631.628712 5419.374349 274.56 2355.7248 20212.11878 173420
40-50 15 45 675 18.58 345.2164 6414.120712 119174.3628 278.7 5178.246 96211.81068 1787615
106 2800 -0.52 16087.7384 -35834.12853 4878477

∑ 𝑓𝑥 2800
Mean = 𝑥̅ = = = 26.42
𝑁 106

The first four central moments are

1 -0.52 1 16087.7384
μ1 = ∑ 𝑓(x − x̅)1 = = −0.005, μ2 = ∑ 𝑓(x − x̅)2 = = 151.77, Standard deviation=√151.77 =
𝑁 106 𝑁 106
12.32
1 -35834.12853 1 4878477
μ3 = ∑ 𝑓(x − x̅)3 = = −338.06, μ4 = ∑ 𝑓(x − x̅)4 = = 46023.37
𝑁 106 𝑁 106
 SKEWNESS

Literally Skewness means ’lack of symmetry’

This helps to have an idea about the shape of the curve of the given data. A distribution is said to be skewed if
1. Mean, Median and Mode are not equal (i.e., they fall at different places)
2. Quartiles are not equidistant from Median
3. The curve will be stretched more towards one side than the other.
𝑀𝑒𝑎𝑛−𝑀𝑜𝑑𝑒
Karl Pearson’s coefficient of skewness = 𝑆𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝑑𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛
KURTOSIS
It shows the convexity of the curve with which we can get more idea about the flatness or peaked Ness of the curve.
𝜇
It is measured by coefficient of coefficient 𝛽2 which is given as 𝛽2 = 𝜇42 or 𝛾2 = 𝛽2 -3
2
1 Find Skewness for the following data:
C.I 0-10 10-20 20-30 30-40 40-50 50-60
f 7 16 12 5 8 3

To find skewness, we calculate Mean, Mode and Standard deviation from the following table:
C.I Midpoints(x) f fx (x-𝑥̅ )^2 f*(x-𝑥̅ )^2

0-10 2800
5 7 35 400
10-20 1600
15 16 240 100
20-30 0
25 12 300 0
30-40 500
35 5 175 100
40-50 3200
45 8 360 400
50-60 2700
55 3 165 900
10800
51 1275

∑ 𝑓𝑥 1275
Mean = 𝑥̅ = = = 25
𝑁 51
𝑓1 −𝑓0
𝑀𝑜𝑑𝑒 = 𝑙 + *c
2𝑓1 −𝑓0 −𝑓2
 16−7
= 10 + *10 =16.923
2∗16−7−12
1 10800
μ2 = ∑ 𝑓(x − x̅)2 = = 211.76,
𝑁 51

Standard deviation=√211.76 = 14.55

𝑀𝑒𝑎𝑛−𝑀𝑜𝑑𝑒 25−16.923
Karl Pearson’s coefficient of skewness = = = 0.555
𝑆𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝑑𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛 14.55

∴ 𝐼𝑡 𝑖𝑠 𝑝𝑜𝑠𝑖𝑡𝑖𝑣𝑒𝑙𝑦 𝑠𝑘𝑒𝑤𝑒𝑑
2 Find kurtosis for the following data
C.I 0-2 2-4 4-6 6-8 8-10

f 23 24 32 15
12

To find kurtosis, we find second and fourth central moments from the following table:

C.I f x=Mid fx (x-𝑥̅ ) (X-𝑥̅ )^2 (X-𝑥̅ )^4 f(X-𝑥̅ )^2 f(X-𝑥̅ )^4
points

0-2 12 1 12 - 18.3184 335.5638 219.8208 4026.765

4.28
2-4 23 3 69 - 5.1984 27.02336 119.5632 621.5373
2.28
4-6 24 5 120 - 0.0784 0.006147 1.8816 0.147517
0.28
6-8 32 7 224 1.72 2.9584 8.752131 94.6688 280.0682
 8-10 15 9 135 3.72 13.8384 191.5013 207.576 2872.52

N=106 560 643.5104 7801.038

∑ 𝑓𝑥 560
Mean = 𝑥̅ = = = 5.283
𝑁 106
1 2 643.5104 1 7801.038
μ2 = ∑ 𝑓(x − x̅) = = 6.07, μ4 = ∑ 𝑓(x − x̅)4 = = 73.594
𝑁 106 𝑁 106
𝜇4 73.594
𝛾2 = 𝛽2 -3= 2 -3 = -3= -1.0029
𝜇2 6.072
∴ It is platykurtic.
 Random Variables

Discrete and continuous r.v s

Discrete Probability Distribution/ Probability Mass
Function/Probability Density Function
Cumulative probability distribution
Expectation
Problems
Random Variable
Definition:
A set whose elements are the numbers assigned to
the outcomes of an experiment.
Usually denoted by uppercase letters such as X
whose elements are denoted by lowercase
𝑥1, 𝑥2, 𝑥3 and so on.
X:1,2,3,4,5,6
 Example 1:x
A coin is tossed twice.
Let X be the random variable representing the number of tails that occur.
Sample space: 𝑆 =𝑇𝑇, 𝑇𝐻,𝐻𝑇, 𝐻𝐻
X:0, 1, 2

Example 2:
Consider an experiment of throwing two dice. We know that this experiment has 36 outcomes .
Let X be the sum of numbers on the uppermost faces.
For Example-The value of X is 2 if outcome of the experiment is (1,1)
Here X is called Random Variable, Where the possible values for X are
X: 2,3,……,12.
 Discrete
Random
Variables
Continuos
Discrete Random Variable
can have finite or infinite number of elements and can be represented by whole numbers.
These values usually arise from counts.
Examples:
1. number of children in a family
2. the Friday night attendance at a cinema
3. the number of patients a doctor sees in one day
4. the number of defective light bulbs in a box of ten
5. the number of “heads” flipped in 3 trials etc.
Continuous Random Variable
infinite number of elements(a range of values) and cannot be represented by whole numbers.
These values usually arise from measurements.

Examples:
1. Waiting time for college bus
2. Length of a random telephone call
3. Amount of coke in a 250 ml pet bottle
4. the amount of sugar in an orange
5. the time required to run a mile etc.
Discrete Probability Distribution/ Probability Mass Function
Let X be a discrete Random Variable taking values as x=0,1,2,3…n, then P(X=x) is called the
probability Mass Function (PMF)/ Probability Distribution function of X if it satisfies the
following:

1. 0 ≤ 𝑃 𝑥 ≤ 1
2. σ 𝑃 𝑥 = 1

x 𝑥1 𝑥2 𝑥3 … 𝑥𝑛
P(X = 𝑝1 𝑝2 𝑝3 … 𝑝𝑛
x)=p(x)

X:0 1 2
P(x):1/4 2/4 1/4
Cumulative Distribution Function:
The cumulative distribution function (cdf) is the probability that the variable takes a value less than or equal to x.
Let X be a R.V (Random Variable) then the cdf associated with X is denoted by F(x) and is defined as
F(x) = P(X ≤ x).
For discrete r.v, it is given as F(𝑥𝑘 ) = σ𝑘𝑖=1 𝑃( 𝑥𝑖 )
Properties:
1. If F is the cdf of a Random Variable X & if a < b then,
P(a < X ≤ b) = F(b)-F(a), P(a ≤ X ≤ b) =P(x=a) + [ F(b)-F(a) ]
P(a < X < b) = [ F(b)-F(a) ] – p(x = b),P(a ≤ X < b) =[ F(b)-F(a) ] – p(x=b)+p(x=a)
Note: if P(x=a) = P(x=b) = 0 then
P(a < X ≤ b) = P(a ≤ X ≤ b) = P(a < X < b) = P(a ≤ X < b) = F(b)-F(a)
2. 0 ≤ F(X) ≤ 1
3. F(x) < F(y) if x<y
Example 1
In the experiment of tossing a coin twice, the four possible outcomes are: HH, HT, TH, and TT. If X is a
random variable representing the number of tails in the outcomes, then X = {0, 1, 2}.

x 0 1 2
P (X = x) P(X=0)=P(0)=1/4 1/2 1/4 E(X)=0*1/4+1*1/2+2*1/4=1

Example2
Find the probability distribution of the random variable X in the experiment of tossing a coin three
times. Let the Random variable X be taken as the number of heads in the outcomes.
S = {HHH, HHT, HTH, HTT, THH, THT, TTH, TTT}

x 0 1 2 3 0,1,2:1/3,1/3,1/3
:(0+1+2)/3=1
P (X = x) 1/8 3/8 3/8 1/8
Mathematical Expectation (Mean)
The expected value of the random variable X is defined as mean which is denoted by E(X) and is
given as
𝐸 𝑋 = 𝜇 = σ𝑛𝑖=1 𝑥𝑖 ∗ 𝑃(𝑥𝑖 )
where p(x) is the probability function of X .
Properties:
1. 𝐸 𝑎 = 𝑎 (where a is constant)
2. 𝐸 𝑎𝑋 = 𝑎 𝐸 𝑋
3. 𝐸 𝑋 ± 𝑌 = 𝐸(𝑋) ± 𝐸(𝑌)
4. 𝐸 𝑋𝑌 = 𝐸 𝑋 ∗ 𝐸(𝑌), where X & Y are independent
5. 𝐸 𝑎𝑋 ± 𝑏 = 𝑎 𝐸(𝑋) ± 𝑏
• Variance
• Variance characterizes the variability in the Distribution and is given as

2
• 𝑉𝑎𝑟 𝑋 = 𝜎 2 = 𝐸(𝑋 − 𝐸(𝑋))2 = 𝐸 𝑋 2 − 𝐸 𝑋

• = σ 𝑥 2 𝑃 𝑥 − 𝜇2

• Properties:

• 1. 𝑉 𝑎 = 0 (where a is constant)
2. 𝑉 𝑎𝑋 = 𝑎2 𝑉 𝑋
3. 𝑉 𝑋 ± 𝑏 = 𝑉(𝑋)
4. 𝑉 𝑎𝑋 ± 𝑏 = 𝑎2 𝑉 𝑋

• Standard deviation: 𝜎= 𝑉𝑎𝑟(𝑋)

Problem: If X denotes the number of heads in a single toss of 4
coins then find (i) P(X<2) (ii) P(1<X≤3) Hhhh,hhht,hhth,ht
Solution: X denotes the number of heads then hh,thhh,httt,thtt,tth
The probability distribution is t,ttth,tthh,hhtt,thth
,htht,thht,htth,tttt,

X 0 1 2 3 4

P(x) P(X=0)=1/ P(X=1)=4/ 6/16 4/16 1/16

16 16

Then
1) 𝑃 𝑋 < 2 = 𝑝 𝑋 = 0 𝑈 𝑋 = 1 = 𝑃(𝐴𝑈𝐵)
=𝑃 𝑋 = 0 + 𝑃 𝑋 = 1
1 4 5
= + =
16 16 16
2) 𝑃 1 < 𝑋 ≤ 3 = 𝑃 𝑋 = 2 + 𝑃 𝑋 = 3
6 4 10
= + =
16 16 16
A box contains 2 red and 4 white balls.2 balls are drawn at random.find expected number of
red balls that are drawn out
Solution: let X: number of red balls that are drawn out,where X is a dicrete r.v
The distribution of X is

X 0 1 2
P(x) 4c2/6c2 2c1*4c1/6c2 2c2/6c2

expected number of red balls that are drawn out

=E(X)=0*4c2/6c2+1*2c1*4c1/6c2+2c2/6c2
E(X)=0*

0*4c2/6c2+1*2c1*4c1/6c2+2c2/6c2
Cumulative Distribution Function
The cumulative distribution function (cdf) of a continuous random variable X is
defined in exactly the same way as the cdf of a discrete random variable.
𝐹(𝑥) = 𝑃(𝑋 ≤ 𝑥).
𝑥
=‫׬‬−∞ 𝑓 𝑥 𝑑𝑥
A continuous random variable X takes on all values in an
interval of numbers. The probability distribution of X is
described by a density curve. The probability of any event is
the area under the density curve and above the values of X that
make up the event.
 
Mean =  =  xf ( x)dx ; −  x  
−


Variance = Var ( X ) =  = x f ( x)dx −  2
2 2

−

M 
1
Median( M ) =
−
 f ( x ) dx = 
M
f ( x ) dx =
2

Mode : Mode is the point or value of X for which f(x) has maximum, if

𝑓 ′ 𝑥 = 0 𝑎𝑛𝑑 𝑓 " 𝑥 < 0

Problem:
Problem:
Problem:
Problem:
Problem:
Problem:
 36
19.9808 = 21.97 - 19.98| = 1.9912
Example 3: Arandom variable Xhas the following probability function
Problem1:

2 3 4 6 7

P) K 2K 2K 3K K 2K' 7K' + K
(i) Determine K (i)Evaluate P(X <6), P(X >6). P(0 <X<5) and PO<X<4)

(ii) if P(X< K) > 2 find the minimum value of Kand, (iv) Determine the distribution
function of X (v) Mean (vi) variance.
JNTU 04S, 05S, 08S,(A) Nov. 11, (H) Dec. 11, (K) May 10S, Nov. 2012, Mar. 2014 (Set No. 2)]
Solution :
7

() Since ), p(x) = 1, we have

x=0

K+ 2K + 2K+ 3K+ K' + 2K' + 7K + K= 1

i.e., 10K' + 9K 1= 0 i.e., (10K- 1) (K+ 1) = 0
1
K= =0.1 (since p(x) >0, so K - 1)
10
(ii) P(X<6) = PX= 0) + P(X =1) +....+ P(X =5)
= 0+ K + 2K + 2K +3K+ K=8K +K'= 0.8 + 0.01 = 0.81
[: K= 0.1]
P(X> 6)=|- P(X <6)= 1-0.81 = 0.19
P(0 <X<5) = P(X= 1)+ PX=2) + P(X = 3) + P(X= 4)
= K+ 2K+ 2K + 3K= 8K= = 0.8
10
P(O <X<4) =P(X =0) +P(X =1) +P(X =2)+P(X =3) +P(X =4)
=0+ K +2K +2K +3K =8K = 8(0.1) = 0.8
Note : P(X <5) =P(X =0) +P(X = )+... +P(X =5)
=0.81 [Refer (i)]
P(X >5)=1-P(X<5) =1-0.81 =0.19
P(0< X <6)= P(X=l)+PX=2) +.... +P(X =5) =0.81
(ii) The required minimum value ofK is obtained as below.

P(X< 1) = P(X= 0) + PX= 1) = 0+K= =0.1

10
P(X < 2) = [P(X = 0) + P(X= 1)]+ P(X= 2)
 48

1 2 3
= 0.3
Probability and s
10 10 10

PX< 3) = [PX= 0) + P(X= ) + PX-2)] +PX -3)

3 0.3 +0.2
P(X< 4) = P(X< 3) + P(X= 4) = 0.5 + = 0.8 > 0.5 =
10
2

.. The ninimum value of K for which P(X < K) > is K = 4

(iv) The distribution function of X is given by the following table :
X F(x) = PX<x)
0
1 K= 1/10
2 3K =3/10
3 5K = 5/10
4 8K = 8/10
5 8K +K?= 81/100
6 8K + 3K' = 83/100
7 9K + 1OK? = 1
7

(v) Mean, u=
i=0
o(0) + 1(K)+ 2(2K) +3(2K) + 4(3K) + 5(K') + 6(2K) + 77K
66 30
66K²+ 30K = =0.66 + 3 = 3.66:K=
100 10 10
7

(vi) Variance =
i=0
K+ 8K+ 18K+ 48K+ 25K? + 72K2 4 343K? + 49K -(3.66
440 124
440K² + 124K - (3.66)= (3.66)²
100 10
4.4 + 12.4 - 13.3956 = 3.4044
 |4K
36 18
Example5:
Problem 2: Arandom variable X has the following probability distribution.
Values of x 0 1 2 3 4 5 6 7 8
P(x) a 3a 5a 7a 9a 1la13a 15a 17a

(i) Determine the value of a.

(ii) Find p(x<3), p(x >3) and p(0<x<5)

(iii) Find the distribution function F(x). |JNTUK) Dec. 2013 (Set No.1))

Solution : () Since p(x;) =l, we have

j=0

a+3a+5a+7a+9a+1la+13a+15a+17a=1
1
i.e., 8la=| 0r a==

81

(ii) p(x<3) = p(r= 0) +p(x =1) +p(x= 2)

9
=a+3a+5a = 9a =
81 9

1 8
P(x23) =l-p(x<3) =1-=

p(0<x< 5) = p(x=1) + p(x=2) + p(x=3) + p(x = 4)

24 8
= 3a+5a +7a+9a = 24a= :
81 27

(iii) The distribution function F(x) of the discrete variate X is defined by

F(x) = P(X < x)= ) p(x) where x is any integer

/=l
 F(X) = P(X<x)
X

a=
0 81

4
4a=
81

9a=
2 81

16
3 16a=
81
25
4 25a =
81
36
5 36a =
81
49
6 49a =
81

64
7 64a =
81

81
8 8la= =1
81
 0.015 +0.02 t 0.03 + 0 + 0.30+ 0.30 + 0.3 = 0.965
of the numbers on the faces
Example 13: A random variable X is defined as the sum
Problem 3:

of X. [JNTU(A) 2009 (Set No. 4)|

when twodice are thrown. Find the mean
when two dice are thrown.x is
Solution: Let x be the sum of the numbers on the faces
is given by
adiscrete random variable whose probability distribution
4 5 6 7 11 12
2 3
5/366/36..2/361/36
P(x;)| 1/362/36 3/364/ 36

.. Mean of X = E() =XP4;

2 3 4
1 + 6/
36 36
+ 4
(36
|+5 36 36

+8/ +10
36 36 (36

22+12)
-(2+6+12+ 20+30 +42+ 40+36+30+
36

2521
36
 46

Example 21 Two dice arethrown. Let Xassign to cach point.(a,

Problem4:

of its numbers i.e. A(a, h) - max.(u, b). Find the probability b) in

Probability S
variable vwith A(s) =(1, 2, 3, 4, 5, 6;. Also find the mean and
JJNTU2004, 2007, 2008S, (A) Dec.
distributiothen.
variance of
X
(OR)Arandom variable Xhas the following distribution? 2009, Nov. istrit
2 3 4 6

P(x) 3 5 7 11
36 36 36 36 36 36

Find (a) the mean (b) variance (c) P(1<x<6) |JNTU (H) Apr.
Solution: The total number of cases are 6 x 6= 36.
The maximum number could be 1,2, 3, 4, 5, 6 i.e., X(s) =
The number 1 will appear only in one case (1, 1). So p(1) =
X(a, b) .012(Sa
1) - P(X=
For maximum 2, favourable cases are (2, 1), (2, 2), (1, 2)
So p(2) =P(X= 2) = 3/36
Pl.l
For maximum 3, favourable cases are (1, 3), (3, 1), (2,
3), (3, 2), (3 2
So p(3) = P(X= 3) = 5/36
For maximum 4, favourable cases are (1,
4),(4, 1),.(2, 4).(4, 2).,(3, 4).4 1
So p(4) = P(X= 4) = 7/36
Similarly, p(5) = P(X= 5)
P((1, 5), (3, 1), (2, 5), (5, 2), (3, 5), (5, 3), (4, 5), (3.4)8
9
36
p(6) = P(X= 6)
P((1, 6), (6, 1), (2, 6), (6,2), (3, 6), (6, 3), (4. 6). (6. 9).6
(6, 5), (6, 6)).
11
36
The required discrete probability distribution is
X= x 2 3 4 5

3 5
PX= x) =Plx) 36 36 36 36
7
36 36
6

() Mean, u = 3 5 7
+ 2. +3.
i=| 36 36 36 +5.1 + 6.
36

(|+ 6 + 15 + 28 + 45 t 66) l61

36 =4.47
36

(ii) Variance, o'-P;

 Functions 47
dom Variables & Distribution
5
)+36 (2)+36 (3' + 7 (4)' + (5' (6-(447Y
36 36 36 36

1
+12 + 45 + I|2 +225 + 396) - (4.47)
791
36
19.9808 = 21.97- 19.981 = 1.9912
 Example 9: From a lot of 10 items containing 3 defectives, asample
Problem 5:
drawn at random. Let the random variable X denote the number of defective

Find the probability distribution of X terns

when the sample is drawn without
(JNTU(K) May replac
sample.

Solution: Obviously X cantake the values 0,

1,2 or 3. 2013 (Se%
Given total number of items = 10
No. of good items = 7
No. of defective items = 3
7! 4!6!
P(X =0) = P(no defective)

P(X= 1)
10CA 4!3!

= P(one defective and 3 good items)

10! 6

3x 7! 4!6!

P(X= 2) =
1°CA 3!4! 10! 2
P(2 defective and 2 good items)
3

10CA 10
Rando
POX= 3) = P(3 defective and I good item)
7 4!

10CA 1°C, 8x9x 10 30

. The probability distribution of random variable X is as follows :
X 1 3

1 3
P(X) 6 10 30

Eromnle
Problem 6:10 : Let X denote the minimum of the two numbers that appear when a pair
Determine thbe
of fair dice is thrown once.
() Discrete probability distribution (i1) Expectation
(iüi) Variance [JNTU2006, (K)Nov. 2009, 2010S (Set No. 1)]
36
Solution: When two dice are thrown, total number of outcomes is 6 x 6 =

(1,1),(1,2), (1,3), (1,4), (1,5), (1, 6)

(2.1),(2.2),(2,3).(2,4), (2,5),(2,6)
(3,1),(3,2), (3,3). (3,4),(3,5),(3, 6)
In this case, sample space S= (4,1),(4, 2). (4,3),(4,4),(4,5), (4,6)|
(5,1),(5,2), (5,3). (5,4),(5, 5), (5,6)
(6.1),(6,2). (6,3), (6,4),(6, 5). (6.,6)
number in S, then the sample
If the random variable X assigns the minimum of its

1 1 1 1 1
1 2 2 2 2 2
1 2 3 3 3 3
space S =
1 2 3 4 4 4
1 2 3 4 5 5
|1 2 3 4 5 6

The minimum number could be 1, 2, 3, 4, 5, 6.

2), (2, 1), (1, 3), (3, 1), (1, 4),
For minimum 1, favourable cases are (1, l), (1,
(4, 1), (1, 5), (5, 1), (1, 6). (6, 1).
11
So PX= 1)
36
(3, 2), (2, 4), (4, 2), (2, 5),
For minimum 2, favourable cases are (2, 2), (2, 3),
(5, 2), (2., 6), (6, 2).
9
So P (X = 2) =
36
54

3). (3, 4). (4. 3). (3. 5). (5, 3). (3, 6
Similarly, PX= 3)= P3, 6). (6,31)
5
PX = 4) = P((4, 4). (4, 5), (5, 4), (4, 6). (6, 4)) = 36
3
PX= 5) = P(S, 5). (5, 6). (6. 5)) = 36

1
PX ==6) = P((6, 6)) = 36

.: The probability distribution is

X 1 2 3 4 5 6

11 9 7 5 3 1
P(X) 36 36 36 36 36 36

(ii) Expectation= Mean =p,x,

11 7 5 3 1
i.e., E(X) = 1.+2. + 3. + 4. + 5. + 6.
36 36 36 36 36 36
1 91
Or (11 + 18 + 21 + 20 + 15 + 6) = =2.5278
36 36
(iii)Variance p, x;-H
11 9
.l+ .4 + 7 .9 +
5
36l6+
3
-.25 + .36
36 36 36 36 36

1
(11 + 36 + 63 + 80 + 75 + 36) 91
36
36
i.e., o' =8.361l -6.3898 = 1.9713
Note: Standard deviation, g = NI.9713= 1404
 items is selected at random from abox o
Example 15 : A sample of 4
Problem 7:
defective items
Find the expected number E of
of which 5are detective.
LJNTU(H) Nov. 2009 (Set No. 4), (A)Nov. 201
among 4 items dr:
Solution: Let Xdenote the number of defective items
items.
4.
Obviously X can take the values 0, 1. 2. 3. or
No. of good items = 7
No. of defective items = 5

P(X= 0) = P(no defective) 7CA 7CA 35 7

12Ca495 495

P(one defective and 3 good items)

99

PX= 1)=

'C, x°C,
X 175 35

12CA 495 99
Random
PX =2) = P(2 defective and 2 good items)
'C, x°C; 210 42
12CA 495 99

PX=3) = P(3 defective and 1 good item)

70 14

12CA 495 99

PX= 4) = P(all are defective)

5Ca
12CA 495

Discrete probability distribution is

99

0 1 2 4
X= X;
7 35 42 14 1
P(X = x) =f() 99 99 99 99 99

Expected number of defective items = E(X) = Exf (x)

7 35 42 14 1 165
= 0. + 1. + 2. +3. + 4.
99 99 99 99 99 99
 Problem 8:
Bxample 19: A player wins if he gets 5 on a single throw of a die, he loses if he gets
A If he wins, he gets Rs. 50, if he loses he gets Rs. 10, otherwise he has to pay
e 15 Find the value of the game to the player. Is it favourable ?
Solution: Range of X= -15, 10, 50:
Sincethere aresix numbers in a die, out of these 5 is only one number, the probability
fgetting Rs. 50is
1
PX=50) =
6
Similarly, the probability of getting two numbers (2 or 4) to win Rs. 10 is
2
PX = 10) =
6 3
is
The probability of getting three numbers (| or 3 or 6) to loose Rs. 15
3 1
PX=- 15) =
6 2
Discrete probability distribution is
-15| 10 50
X=X
1
PX) = p(x) 2 6
25
=(-15).+10.-+50.=Rs.
Expected value of the game = E()= Exp (x;) 2 3

Game is favourable to the player since®E> 0.

Ra

Example22: AAplayer
player tosses two fair coins. He wins Rs. 100/- if head appears,
Problem 9:

200 - if two heads appear. On the other hand he loses Rs.500/- if no head appears.
Rs.
Deterinethe expected value E of the gamne and is the game favourable to the player?
JNTU(K) Nov.2009 (Set No. 3)|
Solution : Let X denote the number of heads occuring in tosses of two fair coins. The
sample space S is
S= (H,T}x (H,T} = (HH, HT, TH, TT;
1
Probability of all 2 leads = P(X =2) =4

Probabilityof all 2tails = P(X =0) = 4

2
Probability of one head =P(X=1)= 4
Discrete probability distribution is
X=x; 1 2

1 2 1
P(x;) 4 4 4
2
Expected value of the. game = 100 x+200 × 500 x
4 4 4

200 + 200- 500 -100

4 4
-=-25 rupees
Game is not favourable to the player since E <0
 Examplel:farandom variable has the probability density f *) as
Problem 10:

2e 2,for x> 0
find the probabilities that it will take on a value
0, for xs0
0.5. |JNTU2001, 20065(Set No. 4)
()between 1 and 3(i) greater than
Solution :
I and 3 is given by
) Theprobability that a variate takes avalue between

P(I<X< 3) = f(r) dt =
1
1
e 2 r)3
= 2 (e -e ) =e2 e
-2
greater than 0.5 is
(i) The probability that avariable takes a value

PX >0.5) = |2e2 dx
0.5 0.5

,-2x
=
2
e
=-le -e)=-(0 -e)el
-2 0.5

is given by
Example 2: If the probability density of a random variable
Problem 11:

k(1-x), for 0<r<l

f) =
0,otherwise
ofkand the probabilities that a random variable having this probability density v
hd the value JJNTU1999
ake on avalue (i) between 0.land 0.2 (ii) greater than 0.5.
 Tka-x). 0 <x <I

Solution: Given f(r) = 0. otherwise

We know d =1
that)

ie.,
0 1

ie., 0+ Jk1-x) dr+0=1

i.e., kx =1 or k
3

3
k =
2
(i) The probability that the variate willtake on a value between 0.1 and 0.
0.2 0.2

P(0.1 <X<0.2) =
0.1 0.1

0.2
=

=
0.008
3

=
0.1-9.007
3 0.2965
(ii) Theprobability that the variate willtake on a
value greater than 0.)

P(X >0.5) =
0.5 0.5

2 0.5 fa-)a0 0.5

-6-04s*). 0.3125
Random

Example 3:12:The probability densityf)ofacontinuous random variable is given by

Problem
fu) =ce I<0. Show that c 1/2 and find that the mean and variance of the
distribution. Also find the probability that the variate lies between Oand 4.
JJNTU Jan. 2007,(A) Nov. 2010 (Set No. 2), (K) May 2013 (Set No. 4)|
Solution : Given f()= ce

We have f)a=l [sincethe total probability is unityl

i.e.. [celàt =l ie., c ed-1

-0

ie. 2c Je"d =1:: 1s an even

function
i.e., 2c e d =1 [: in 0 <x<0, Jx|=x]
0

2e(-e-l - 2c (0- 1) ’ 2c = 1 .. c=
2

1
Hence f(r) = ce =2
() Mean of the distribution,

2
xed =0, since integrand is odd.
(i) Variance of the distribution,

= 2. 2 xe d, since integrand is even

0
 r2e e e
probability ang
rea -1
-2x-
1
-+2
-1

= [0- (- 2)]= 2
(ii)The probability btween 0 and 4 = P(0<X<4)
4 4
1 1
de
2 2

|:: in 0<I<t,
1
2 (et-)
2 (1 -e) =0.4908 (nearly)
 function
Example5: A continuous random vaiable has the probability density
Problem 14:

kxer for x > 0,2>0

f(r) = |0,otherwise

Determine (i) k (ii)Mean (iii) Variance

|JNTU(A)Dec. 2009, Nov. 2010, Dec. 2011, (H) May 2011, Nov. 2012, (K)) May 2013 (Set No. ))
Solution : (i) Since the total probability is unity, we have
76
ty and

i.e.,

=|
i.e..

i.e., ao-0-o|-tok-2
Now f () becomes
a' xe -Ax for x > 0, >0
f(x) = 0, otherwise

(ii) Mean of the distribution, =

i.e., u =
0

2
0-0-0)-0-0
(i) Variance of the distribution, o² = f ) dt -u
- 00

4
i.e., g=
0

=
+6x

0-00-09-fo-0-09)
6 4 2
Problem 15: Acontinuous random variable
Example 6: Nis defined by

f)=l6 (6-2v). if 1srsl

0if Isxs3
0. elsewhere

Verify that f (r) is adensity function and also find the mean of X.
|JNTU2003S, (K) May 2013 (Set No. 3)]
Solution :f(r) is clcarly >0for every r in (-3, 3] and
-3 -1

-1
3

1 (3+x)°| 1(3-r) 7
3 2
16 J-3 16 16 -3

1 1
48
(8-0)+ 8 6x
3 48
(0-8)

+
6 =1

Hence the function f (x) satisfies the requirements for a density function.

Mean of f() = xf) dr

16 16
-3
 I6
3

[. the integrand of the second integral is

3

16
for+6r+rd 16
3

9r? 6r3
9r? 6x
+
16 2 3 4
l6 2 3 4
3

81

16
G-)-)(G-2»4
54 --54 + +
16
 Example 12: For the continuous random variable X whose probability
devitey
Problem 16:
|Cx2 ). if 0<rS2
Is given by where c is aconstant
0.othervwise
Find c. mean and variance of X. JNTU 2003,244
(OR) The frequency function of a continuous random variable
fx)=Vox (2-x),0<x<2. Find the value of o, mean and variance of X
|JNTU(A) Dec.
.20W9 (Sa
Solution: ()) Since the total probability is unity, we have

|f) dx =1. So () dr =1

2 .3

i.e.. lex (2-x) de =1I i.e.,cx? =]

0
3

8 3
i.e., C
(+-})-1
4
i.e.,
4c
3
= ] orc=
4

3x
4
(2-x), if 0<x<2
0, otherwise

(ii) Mean of X=EX)= x f ( ) dr

 2
3r 3
ie., = T2-) dr =4 (2r-)dr

3 2r3 2* 12
4 4 3 4 = ]
12

2
3x
(ii) Variance of X= V) = xfr) d -u = 2-x) dr -(I)'
0

3
2
4 4 4

3 32 32

Problem 17:
4 4 |-24-1--!
Example 13: For the continuous probability function f(r) = kxe when x> 0, find
() k (ii) Mean (iii) Variance |JNTU 2005, 2005S, 2007, (K) Dec. 2013 (Set No. 4)|

Solution :() We havef) dx=|

|kr'e "dr =1 (::x> 0)

0

i.e., k-e")-2x (e") +2(-e")

ie, k-e +2x +2)), = 1
i.e., k (0 + 2) = 1 or k=
2

(ii) Mean =

kr'-e)- 3r'(e')+ 6x(-e")- 6e

k-er'+3x' +6x +6), -k [0 t 6<= 6k
. =
Joint Probability Distributions
In general, if 𝑋, 𝑌, … … 𝑃 are n random variables, the probability distribution that defines their simultaneous
behavior is called a joint probability distribution. When there are only two random variables of interest, we also
use the term bivariate probability distribution or bivariate distribution to refer to the joint distribution.

Types:
Joint probability mass function of the discrete random variables X and Y {P(x, y)}
Joint Probability density Function of the continuous random variables X and Y {f(x, y)}
Examples for discrete r.v s
• Price Vs. demand taken
• Year in college vs. Number of credits taken
Examples for continuous r.v s
• Dosage of a drug (ml) vs. Blood compound measure (percentage)
Bivariate Distributions
For both discrete and continuous random variables we will discuss the following...
• Joint Distributions (for two or more r.v s)
• Marginal Distributions (computed from a joint distribution)
• Conditional Distributions (e.g. P(Y = y/X = x))
• Joint cumulative distribution function
• Independence for r.v s X and Y
Joint probability mass function
𝐲𝟏 𝐲𝟐 𝐲𝐣 𝐲𝐧 𝑷𝑿 𝒙 = 𝑷 𝑿 = 𝒙
X\Y = ෍ 𝑷(𝒙, 𝒚)
𝒚

𝐱𝟏 P(𝑥1 , 𝑦1 ) P(𝑥1 , 𝑦2 ) P(𝑥1 , 𝑦j ) P(𝑥1 , 𝑦𝑛 ) 𝑷𝑿 𝑥1 = 𝑷 𝑿 = 𝑥1

= ෍ 𝑷(𝑥1 , 𝒚)
𝒚

𝐱𝟐 P(𝑥2 ,𝑦1 ) P(𝑥2 , 𝑦2 ) P(𝑥2 , 𝑦j ) P(𝑥2 , 𝑦n )

𝐱𝐢 𝑃(𝑥𝑖 , 𝑦1 ) 𝑃(𝑥𝑖 , 𝑦2 ) P(𝑥𝑖 , 𝑦𝑗 ) 𝑃(𝑥𝑖 , 𝑦𝑛 ) 𝑷𝑿 𝑥i = 𝑷 𝑿 = 𝑥i

= ෍ 𝑷(𝑥i , 𝒚)
𝒚

𝐱𝐦 P(𝑥𝑚 , 𝑦1 ) P(𝑥𝑚 , 𝑦2 ) P(𝑥𝑚 , 𝑦𝑗 ) P(𝑥𝑚 , 𝑦𝑛 )

𝑷𝒀 𝒚 = 𝑷 𝒀 = 𝒚 𝑷𝒀 𝑦𝑗 = 𝑷 𝒀 = 𝑦𝑗
෍ ෍ 𝑃𝑋,𝑌 𝑥, 𝑦 = 1
= ෍ 𝑷(𝒙, 𝒚) = ෍ 𝑷(𝒙, 𝑦𝑗 )
𝒙 𝒙
Joint probability mass function
The joint probability mass function of the discrete random variables X and Y is denoted as 𝑃𝑋,𝑌 𝑥, 𝑦 = 𝑃(𝑋 = 𝑥, 𝑌 = 𝑦),
and it satisfies
(1) 0 ≤ 𝑃𝑋,𝑌 𝑥, 𝑦 ≤1 and (2) σ σ 𝑃𝑋,𝑌 𝑥, 𝑦 = 1

Marginal Probability Distributions

If we are given a joint probability distribution for X and Y , we can obtain the individual probability distributions for X or
for Y which are also called the Marginal Probability Distributions. If X and Y are discrete random variables with joint
probability mass function 𝑃𝑋,𝑌 (𝑥, 𝑦), then the marginal probability mass functions of X and Y are
𝑃𝑋 𝑥 = 𝑃 𝑋 = 𝑥 = σ𝑦 𝑃(𝑥, 𝑦) and 𝑃𝑌 𝑦 = 𝑃 𝑌 = 𝑦 = σ𝑥 𝑃(𝑥, 𝑦)

Joint Cumulative Distribution function

It is represented by F(x,y) and is given as 𝐹 𝑥, 𝑦 = P X ≤ 𝑥, 𝑌 ≤ 𝑦 = σ𝑥 σ𝑦 𝑃(𝑥, 𝑦)

Test for independence

Two r.vs are said to be independent iff the joint probability is equal to the product of marginal probabilities.
𝑃(𝑥, 𝑦) = 𝑃(𝑥) ∗ 𝑃(𝑦), 𝑓𝑜𝑟 𝑎𝑙𝑙 X 𝑎𝑛𝑑 Y
Illustration: Let X and Y be the numbers sided up when a pair of dice is thrown, the joint probability can
be shown as follows:

X\Y 𝟏 𝟐 3 4 𝟓 6 𝑃𝑋 𝑥 = 𝑃 𝑋 = 𝑥
= ෍ 𝑃(𝑥, 𝑦)
𝑦

𝟏 P(1,1) =1/36 P(1,2) =1/36 P(1,3) =1/36 P(1,4)=1/36 P(1,5)=1/36 P(1,6)=1/36 𝑃𝑋 1 =σ𝑦 𝑃(1, 𝑦)=1/6

𝟐 P(2,1) =1/36 P(2,2) =1/36 P(2,3) =1/36 P(2,4) =1/36 P(2,5) =1/36 P(2,6) =1/36 𝑃𝑋 2 =1/6

3 P(3,1) =1/36 P(3,2) =1/36 P(3,3) =1/36 P(3,4) =1/36 P(3,5) =1/36 P(3,6) =1/36 𝑃𝑋 3 =1/6

4 P(4,1) =1/36 P(4,2) =1/36 P(4,3) =1/36 P(4,4) =1/36 P(4,5) =1/36 P(4,6) =1/36 𝑃𝑋 4 =1/6

𝟓 P(5,1) =1/36 P(5,2) =1/36 P(5,3) =1/36 P(5,4) =1/36 P(5,5) =1/36 P(5,6) =1/36 𝑃𝑋 5 =1/6

6 P(6,1) =1/36 P(6,2) =1/36 P(6,3) =1/36 P(6,4) =1/36 P(6,5) =1/36 P(6,6) =1/36 𝑃𝑋 6 =1/6

𝑃𝑌 𝑦 𝑃𝑌 1 =σ𝑥 𝑃(𝑥, 1) 𝑃𝑌 2 =1/6 𝑃𝑌 3 =1/6 𝑃𝑌 4 =1/6 𝑃𝑌 5 =1/6 𝑃𝑌 6 =1/6

=𝑃 𝑌=𝑦 =1/6
= ෍ 𝑃(𝑥, 𝑦)
𝑥
Problem1:Consider two r.vs with X∈{0,1,2} and Y∈{1,2,4).The following table gives the joint probability mass function:

X\Y 1 2 4
0 0.1 0.1 0
1 0.2 0.1 0.1
2 0 0.2 0.2

Calculate i)E(X), ii)E(Y), iii)Conditional density function of X given Y, iv) F(1,4) and v) test whether X and Y are independent or not
Solution: From the given joint probability distribution, we calculate the marginal probabilities of X & Y as follows:
X\Y 1 2 4 𝑷𝑿 𝒙
0 0.1 0.1 0 𝑃𝑋 0 =0.2

1 0.2 0.1 0.1 𝑃𝑋 1 =0.4

2 0 0.2 0.2 𝑃𝑋 2 =0.4

𝑷𝒀 𝒚 𝑃𝑌 1 =0.3 𝑃𝑌 2 =0.4 𝑃𝑌 4 =0.3

෍ ෍ 𝑃𝑋,𝑌 𝑥, 𝑦 = 1

Contd….
 Marginal distribution of Y is
Marginal distribution of X is
Y 1 2 4
X 0 1 2
𝑷𝒀 𝒚 𝑃𝑌 1 =0.3 𝑃𝑌 2 =0.4 𝑃𝑌 4 =0.3
𝑷𝑿 𝒙 𝑃𝑋 0 =0.2 𝑃𝑋 1 =0.4 𝑃𝑋 2 =0.4

i)𝐸 𝑋 = σ 𝑥 ∗ 𝑝 𝑥 =0 ∗ 0.2 + 1 ∗ 0.4 + 2 ∗ 0.4 = 1.2

ii) 𝐸(𝑌) = σ 𝑦 ∗ 𝑝(𝑦)=1 ∗ 0.3 + 2 ∗ 0.4 + 4 ∗ 0.3 = 2.3
𝑝(𝑥∩𝑦)
iii)Conditional density of X given Y is P(X/Y) which can be calculated as 𝑃 𝑋Τ𝑌 = 𝑝(𝑦)

𝑋=0Τ 𝑝(0,1) 0.1 𝑋=2Τ 𝑝(2,4) 0.2

EX : 𝑃 𝑌=1 = = 0.3=0.333,……… 𝑃 𝑌=4 = = 0.3=0.667
𝑝𝑌 (1) 𝑝𝑌 (4)

𝑌=2Τ 𝑝(1,2)
𝑐𝑜𝑛𝑑𝑖𝑡𝑖𝑜𝑛𝑎𝑙 𝑑𝑒𝑛𝑠𝑖𝑡𝑦 𝑓𝑜𝑟 𝑌 𝑔𝑖𝑣𝑒𝑛 𝑋 𝑖𝑠 𝑃 𝑋=1 =𝑝 and so on..
𝑋 (1)

iv)𝐹 1,4 = σ1𝑋=0 σ4𝑌=1 𝑝(𝑥, 𝑦) = 0.1 + 0.1 + ⋯ … + 0.2 + 0.2 = 0.6
v)Test for independence is 𝑃(𝑥, 𝑦) = 𝑃(𝑥) ∗ 𝑃(𝑦), 𝑓𝑜𝑟 𝑎𝑙𝑙 𝑥 𝑎𝑛𝑑 𝑦
Consider 𝑃𝑋 0 *𝑃𝑌 1 =0.2*0.3=0.06≠ 𝑝 0,1
∴ 𝑋 & 𝑌 𝑎𝑟𝑒 𝑛𝑜𝑡 𝑖𝑛𝑑𝑒𝑝𝑒𝑛𝑑𝑒𝑛𝑡
Problem 2:
Suppose that 2 batteries are randomly chosen from the following group of 12 batteries: 3 new 4 used (working) 5 defective. Let X
denote the number of new batteries chosen. Let Y denote the number of used batteries chosen. Compute i) joint probability distribution
of X&Y, ii)Mean number of used batteries chosen, iii)Test whether the selection of new batteries is influencing the selection of used
batteries or not.
Solution: Total number of batteries=12
X: number of new batteries chosen and Y: number of used batteries chosen
The joint probability distribution can be written as follows:

X\Y 0 1 2 Marginal Probabilities of X

0 5 4 5
∗ 4 𝑃𝑋 0 = 0.272
2 1 1
= = 0.151 = = 0.303
12 12 = 2 = 0.091
2 2 12
2
1 3 5 4 3 0 𝑃𝑋 1 =0.409
∗ ∗
1 1 1 1
= 12 = 0.227 = 12 = 0.182
2 2

2 3 0 0 𝑃𝑋 2 = .046
2
= 12 = 0.046
2

Marginal 𝑃𝑌 0 = 0.424 𝑃𝑌 1 = 0.485 𝑃𝑌 2 = 0.091

Probabilities of Y ෍ ෍ 𝑃𝑋,𝑌 𝑥, 𝑦 = 1
 Marginal distribution of Y is
Marginal distribution of X is
X 0 1 2 Y 0 1 2
𝑷𝑿 𝒙 𝑃𝑋 0 =0.272 𝑃𝑋 1 =0.409 𝑃𝑋 2 =0.046 𝑷𝒀 𝒚 𝑃𝑌 0 =0.424 𝑃𝑌 1 =0.485 𝑃𝑌 2 =0.091

Then,
i) Mean number of used batteries chosen= 𝐸(𝑌) = σ 𝑦 ∗ 𝑝(𝑦)
= 0 ∗ 0.424 + 1 ∗ .485 + 2 ∗ 0.091 = 0.667 ≅ 1 𝑏𝑎𝑡𝑡𝑒𝑟𝑦
ii)Test for independence:
Consider 𝑃𝑋 0 *𝑃𝑌 0 =0.1153≠ 𝑝 0,0
∴ 𝑋 & 𝑌 𝑎𝑟𝑒 𝑛𝑜𝑡 𝑖𝑛𝑑𝑒𝑝𝑒𝑛𝑑𝑒𝑛𝑡
 X&Y are two discrete r.vs X&Y are two continuous r.vs
𝐽𝑜𝑖𝑛𝑡 𝑝𝑟𝑜𝑏𝑎𝑏𝑖𝑙𝑖𝑡𝑦 𝐽𝑜𝑖𝑛𝑡 𝑝𝑚𝑓 = 𝑃𝑋,𝑌 𝑥, 𝑦 𝐽𝑜𝑖𝑛𝑡 𝑝𝑑𝑓 = 𝑓(𝑥, 𝑦)
𝑑𝑖𝑠𝑡𝑟𝑖𝑏𝑢𝑡𝑖𝑜𝑛 𝑓𝑢𝑛𝑐𝑡𝑖𝑜𝑛
𝑀𝑎𝑟𝑔𝑖𝑛𝑎𝑙 𝑃𝑟𝑜𝑏𝑎𝑏𝑖𝑙𝑖𝑡𝑦 𝑚𝑎𝑟𝑔𝑖𝑛𝑎𝑙 𝑝𝑟𝑜𝑏𝑎𝑏𝑖𝑙𝑖𝑡𝑦 𝑚𝑎𝑠𝑠 𝑓𝑢𝑛𝑐𝑡𝑖𝑜𝑛𝑠 𝑚𝑎𝑟𝑔𝑖𝑛𝑎𝑙 𝑝𝑟𝑜𝑏𝑎𝑏𝑖𝑙𝑖𝑡𝑦 𝑑𝑒𝑛𝑠𝑖𝑡𝑦
𝑓𝑢𝑛𝑐𝑡𝑖𝑜𝑛𝑠
𝑜𝑓 𝑋 𝑎𝑛𝑑 𝑌 𝑎𝑟𝑒 𝑓𝑢𝑛𝑐𝑡𝑖𝑜𝑛𝑠 𝑜𝑓 𝑋 𝑎𝑛𝑑 𝑌 𝑎𝑟𝑒
𝑃𝑋 𝑥 = 𝑃 𝑋 = 𝑥 = σ𝑦 𝑃(𝑥, 𝑦) +∞
𝑓 𝑥 = ‫׬‬−∞ 𝑓 𝑥, 𝑦 𝑑𝑦

+∞
𝑃𝑌 𝑦 = 𝑃 𝑌 = 𝑦 = ෍ 𝑃(𝑥, 𝑦) 𝑓 𝑦 =න 𝑓 𝑥, 𝑦 𝑑𝑥
𝑥 −∞

𝑥 𝑦 𝑥 𝑦
𝐶𝑢𝑚𝑢𝑙𝑎𝑡𝑖𝑣𝑒 𝑑𝑖𝑠𝑡𝑟𝑖𝑏𝑢𝑡𝑖𝑜𝑛
𝑓𝑢𝑛𝑐𝑡𝑖𝑜𝑛 𝐹 𝑥, 𝑦 = ෍ ෍ 𝑃(𝑥, 𝑦) 𝐹 𝑥, 𝑦 = න න 𝑓 𝑥, 𝑦 𝑑𝑦𝑑𝑥
−∞ −∞
𝐼𝑛𝑑𝑒𝑝𝑒𝑛𝑑𝑒𝑛𝑐𝑒 𝑜𝑓 𝑟. 𝑣𝑠 𝑃(𝑥, 𝑦) = 𝑃(𝑥) ∗ 𝑃(𝑦), 𝑓𝑜𝑟 𝑎𝑙𝑙 𝑥 𝑎𝑛𝑑 𝑦 𝑓(𝑥, 𝑦) = 𝑓(𝑥) ∗ 𝑓(𝑦)
Joint probability density function:
A function f(x,y) is said to be joint pdf of two r.v s X&Y, where −∞ ≤ 𝑋 ≤ ∞ 𝑎𝑛𝑑
−∞ ≤ 𝑌 ≤ ∞ ,iff
1. f(x,y)≥ 0 and ≤1
∞ ∞
2. ‫׬‬−∞ ‫׬‬−∞ 𝑓 𝑥, 𝑦 𝑑𝑦𝑑𝑥 =1

𝑚𝑎𝑟𝑔𝑖𝑛𝑎𝑙 𝑝𝑟𝑜𝑏𝑎𝑏𝑖𝑙𝑖𝑡𝑦 𝑑𝑒𝑛𝑠𝑖𝑡𝑦 𝑓𝑢𝑛𝑐𝑡𝑖𝑜𝑛𝑠 𝑜𝑓 𝑋 𝑎𝑛𝑑 𝑌 𝑎𝑟𝑒

+∞
𝑓 𝑥 = ‫׬‬−∞ 𝑓 𝑥, 𝑦 𝑑𝑦

+∞
𝑓 𝑦 =න 𝑓 𝑥, 𝑦 𝑑𝑥
−∞

Joint cumulative distribution function of X &Y is represented as F(x,y) which is given as

𝐹 𝑥, 𝑦 = P X ≤ 𝑥, 𝑌 ≤ 𝑦 =

𝑥 𝑦

න න 𝑓 𝑥, 𝑦 𝑑𝑦𝑑𝑥
−∞ −∞
𝐼𝑛𝑑𝑒𝑝𝑒𝑛𝑑𝑒𝑛𝑐𝑒 𝑜𝑓 𝑟. 𝑣𝑠:
Two r. v s are said to be independent iff the joint density is equal to the product
of marginal densities.
𝑓 𝑥, 𝑦 = 𝑓 𝑥 ∗ 𝑓 𝑦

Conditional density functions:

The conditional probability density function for X given Y is 𝑓 𝑥Τ𝑦 𝑎𝑛𝑑 𝑖𝑠 𝑔𝑖𝑣𝑒𝑛 𝑎𝑠

𝑓(𝑥, 𝑦)
𝑓 𝑥ൗ𝑦 =
𝑓𝑌 (𝑦)
similarly
𝑦 𝑓(𝑥, 𝑦)
𝑓 ൗ𝑥 =
𝑓𝑋 (𝑥)
 X&Y are two discrete r.vs X&Y are two continuous r.vs
𝐽𝑜𝑖𝑛𝑡 𝑝𝑟𝑜𝑏𝑎𝑏𝑖𝑙𝑖𝑡𝑦 𝐽𝑜𝑖𝑛𝑡 𝑝𝑚𝑓 = 𝑃𝑋,𝑌 𝑥, 𝑦 𝐽𝑜𝑖𝑛𝑡 𝑝𝑑𝑓 = 𝑓(𝑥, 𝑦)
𝑑𝑖𝑠𝑡𝑟𝑖𝑏𝑢𝑡𝑖𝑜𝑛 𝑓𝑢𝑛𝑐𝑡𝑖𝑜𝑛
𝑀𝑎𝑟𝑔𝑖𝑛𝑎𝑙 𝑃𝑟𝑜𝑏𝑎𝑏𝑖𝑙𝑖𝑡𝑦 𝑚𝑎𝑟𝑔𝑖𝑛𝑎𝑙 𝑝𝑟𝑜𝑏𝑎𝑏𝑖𝑙𝑖𝑡𝑦 𝑚𝑎𝑠𝑠 𝑓𝑢𝑛𝑐𝑡𝑖𝑜𝑛𝑠 𝑚𝑎𝑟𝑔𝑖𝑛𝑎𝑙 𝑝𝑟𝑜𝑏𝑎𝑏𝑖𝑙𝑖𝑡𝑦 𝑑𝑒𝑛𝑠𝑖𝑡𝑦
𝑓𝑢𝑛𝑐𝑡𝑖𝑜𝑛𝑠
𝑜𝑓 𝑋 𝑎𝑛𝑑 𝑌 𝑎𝑟𝑒 𝑓𝑢𝑛𝑐𝑡𝑖𝑜𝑛𝑠 𝑜𝑓 𝑋 𝑎𝑛𝑑 𝑌 𝑎𝑟𝑒
𝑃𝑋 𝑥 = 𝑃 𝑋 = 𝑥 = σ𝑦 𝑃(𝑥, 𝑦) +∞
𝑓 𝑥 = ‫׬‬−∞ 𝑓 𝑥, 𝑦 𝑑𝑦

+∞
𝑃𝑌 𝑦 = 𝑃 𝑌 = 𝑦 = ෍ 𝑃(𝑥, 𝑦) 𝑓 𝑦 =න 𝑓 𝑥, 𝑦 𝑑𝑥
𝑥 −∞

𝑥 𝑦 𝑥 𝑦
𝐶𝑢𝑚𝑢𝑙𝑎𝑡𝑖𝑣𝑒 𝑑𝑖𝑠𝑡𝑟𝑖𝑏𝑢𝑡𝑖𝑜𝑛
𝑓𝑢𝑛𝑐𝑡𝑖𝑜𝑛 𝐹 𝑥, 𝑦 = ෍ ෍ 𝑃(𝑥, 𝑦) 𝐹 𝑥, 𝑦 = න න 𝑓 𝑥, 𝑦 𝑑𝑦𝑑𝑥
−∞ −∞
𝐼𝑛𝑑𝑒𝑝𝑒𝑛𝑑𝑒𝑛𝑐𝑒 𝑜𝑓 𝑟. 𝑣𝑠 𝑃(𝑥, 𝑦) = 𝑃(𝑥) ∗ 𝑃(𝑦), 𝑓𝑜𝑟 𝑎𝑙𝑙 𝑥 𝑎𝑛𝑑 𝑦 𝑓(𝑥, 𝑦) = 𝑓(𝑥) ∗ 𝑓(𝑦)
Covariance: It is the mean of product of deviations of two variables taken from their respective
means.

It is given as

Covariance (X, Y) = cov(X, Y) = E((𝑋 − 𝐸(𝑋)) ∗ (𝑌 − 𝐸(𝑌)))

1
= ∑(𝑥 − 𝑥̅ ) ∗ (𝑦 − 𝑦̅ )
𝑛
(or)

= 𝐸(𝑋𝑌) − 𝐸(𝑋) ∗ 𝐸(𝑌)

1
= ∑ 𝑥𝑦 − 𝑥̅ 𝑦̅
𝑛
Note: Covariance is zero for independent random variables