100% found this document useful (3 votes)
1K views10 pages

DA1

The document contains statistics problems involving concepts such as mean, median, mode, variance, standard deviation, probability distributions, moments, Pearson's coefficients, and joint probability distributions. It includes 8 questions with solutions calculating and analyzing these statistical metrics for various frequency distributions and probability density functions.

Uploaded by

SPAM rohit
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
100% found this document useful (3 votes)
1K views10 pages

DA1

The document contains statistics problems involving concepts such as mean, median, mode, variance, standard deviation, probability distributions, moments, Pearson's coefficients, and joint probability distributions. It includes 8 questions with solutions calculating and analyzing these statistical metrics for various frequency distributions and probability density functions.

Uploaded by

SPAM rohit
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 10

16BCE0766

STATISTICS FOR
ENGINEERS
DIGITAL ASSIGNMENT - I

NAME: ROHIT VEMPARALA


REGISTRATION NUMBER: 16BCE0766
SLOT: E1+TE1
FACULTY: EASWARAMOORTHY
16BCE0766

QUESTIONS

1. Compute the mean, median and mode for the following frequency distribution.

Solution

CLASS INTERVAL FREQUENCY Xi Di CF Fidi


20-25 30 22.5 -4 30 -120
25-30 160 27.5 -3 190 -480
30-35 210 32.5 -2 400 -420
35-40 180 37.5 -1 580 -180
40-45 145 42.5 0 725 0
45-50 105 47.5 1 830 105
50-55 70 52.5 2 900 140
55-60 60 57.5 3 960 180
60-65 40 62.5 4 1000 160

A (Assumed mean) = 42.5 C (class width) = 5


f = N = 1000 di = (xi - A )/c
fi di = -615

Mean

Mean = A + (( fi di)/N)* c = 42.5 + (-615/1000) = 39.45

Median

N = 1000 Here N is even


N/2, (N+1)/2 = 500,500.5 Average of N/2 and (N+1)/2 = 500.25
Median class = 35*40

l = 35 f = 180 m = 400

Median = l + ((N/2 - m)/N)*c = 35=50/18 = 37.7

Mode

F1 = 210 f0 = 160 f2 = 180 l=30 c=5

mode = l + ((f1 f0)/(2*f1+(f0+f2))) * c = 30 + (210-160)/(420-340)*5 = 33.125


16BCE0766

2. Calculate the mean, variance and standard deviation for the following frequency
Distribution, and hence obtain the value of co-efcient of variation.

Solution

Size (x) Frequency (f)


0 1
1 8
2 28
3 56
4 70
5 56
6 28
7 8
8 1

f = N = 256 fixi = 1024

Mean = (fixi)/N = 1024/256 = 4

Standard Deviation =

^
( )
= 1.4142
Variance = 2

Coefficient of variation = 100 = 35.3553

3. Find the rst four moments about the mean and Pearsons co-efficient 1 and 2 for the
following frequency distribution, and also comment the nature of skewness and kurtosis
of the distribution based on the Pearsons co-efficient 1 and 2

Solution
Class f x d fd
170-180 52 175 -4 -208 832 -3378 13312
180-190 68 185 -3 -204 612 -1836 5508
190-200 85 195 -2 -170 340 -680 1360
200-210 92 205 -1 -92 92 -92 92
210-220 100 215 0 0 0 0 0
220-230 95 225 1 95 95 95 95
230-240 70 235 2 140 280 560 420
240-250 28 245 3 84 252 756 2268
Total 590 -355 2503 -4525 2375
16BCE0766
355
1 = = 10 = 6.0169
590
2503
2 = = 100 = 424.2372
590
4525
3 = = 1000 = 7669.4912
590
23755
4 = = 10000 = 402627.1186
590

1 = 0
2 = 2 (1 ) = 388.0341
3 = 3 3(2 1) + 2(1 ) = 229.5431
4 = 4 4(3 1 ) + 62 (1 ) 3(1 ) = 214108.882

Pearsons Coefficients 1 and 2


( ) .
1 = = .
= 0.4611

.
2 = ( )
= .
= 1.4219

Nature of kurtosis is platykurtic

Coefficient of skewness
1(2 + 3) 1.4219 + 3 3.0026
= = 04611 = = 0.3223
2(5(2) 61 9) 2(7.1095 2.7666) 9 9.3142

Nature of skewness is negatively skewed

4. Find the value of a, P(X < 3), cumulative distribution function, mean, variance and
standard deviation of the discrete random variable (X) with the following probability
distribution.

P(X=xi) = 1 => a+3a+5a+7a+9a+11a+13a+15a=17a = 1

81a = 1
a = 1/81

P(X < 3) = P(X=0) + P(X=1) + P(X=2) = a+3a+5a = 9a = 1/9

Cumulative distribution function CDF (F(x)) = P (X<=x) = P(x)

F(0) = 0 F(1) = P(X = 0) = 1/81

Similarly
F (2) = 4/81 F (3) = 1/9 F(4) = 16/81 F(5) = 25/81 F(6) = 36/81 F(7) = 49/81
F (8) = 64/81 F(9) = 1
16BCE0766
Mean = x*P(x) = 0 + 1(3/81) + 2*(5/81) +3*(7/81) +4*(9/81) + 5*(11/81) + 6*(13/81) +
7*(15/81) + 8*(17/81) = 444/81 = 5.481


Variance = (x-mean)^2 * P(x) or () = 34.5185 (5.4814)

= ( 5.48)^2*1/81 +

Variance = 4.47277
Standard Deviation SD = = 2.1148

5. Find the value of k, cumulative distribution function, mean, variance and standard
deviation of the continuous random variable (X) with the following probability density
function.

Solution
() = 1
(2 ) + 0 = 1
2 = 1
k=
Cumulative distribution function

() = 1 F(x) = () for a value of x

F (x < 0) = 0
F (0 < x < 2) = () = 1
F (x > 2) = 0

Mean = () = (2 ) = = 1

Variance = ( ) () = ( 1) ()(2 ) = =

Standard deviation = = 1/2

6. Three balls are drawn at random without replacement from a box containing 2 white, 3
red and 4 black balls. If X denotes the number of white balls drawn and Y denotes the
number of red balls drawn, nd the joint probability distribution of (X; Y ).

Solution:

2 White balls, 3 Red balls, 4 black balls and 3 balls are drawn at random from the given balls

X takes the values 0, 1, 2


Y takes the values 0, 1,2 3
16BCE0766
The different cases are as follows

P(X=0 Y=0) = 4C3/9C3 = 1/21 P(X=0 Y=2) = (3C2 X 4C1)/9C3 = 1/7


P(X=0 Y=1) = (3C1 X 4C2)/9C3 = 3/14 P(X=0 Y=3) = 1/84
P(X=1 Y=0) = 1/7 P(X=1 Y=2) = 1/14
P(X=1 Y=1) = 2/7 P(X=1 Y=3) = 0
P(X=2 Y=0) = 1/14 P(X=2 Y=2) = 0
P(X=2 Y=1) = 1/28 P(X=2 Y=2) = 0
P(X=2 Y=3) = 0

Joint distribution Table

X Y
0 1 2 3
0 1/21 3/14 1/7 1/84
1 1/7 2/7 1/14 0
2 1/21 1/28 0 0

7. The joint probability density function of two-dimensional continuous random variable (X;
Y ) is given by

Solution

= 1

Substitution method
Taking = => 2xdx=dt => xdx = dt and = => 2ydy=dk => ydy = dk
K(0-1)(0-1) = 4
K=4

f(x,y) = 4xy

Marginal Functions

f (x) = 4 = 2 f (y) = 4 = 2

f(x) . f(y) = 2 . 2 = 4 = (, )

Hence x and y are independent from the property

Hence proved

8. A line of length a units is divided into two parts. If the rst part is of length X, then find
E(X), Var(X), and E [X(a X)].
Solution:
X is uniformly distributed in (0,0)
16BCE0766
f(x) = 1/a
E(x) = () = =

( ) = () =
3
Variance (x) = ( ) () = =

( ) = () ( ) = =
3 4 12

9. Compute the coefcient of correlation between the following two set of measures X and
Y and hence comment the nature of correlation between X and Y.

Solution

Changing the origin

Minimum value among the values of X is 65 Minimum value among the values of Y is 64
Maximum value among the values of X is 71 Maximum value among the values of Y is 72
New origin of X = (65 + 71)/2 = 68 New origin of Y = (64 + 72)/2 = 68

X=xi Y=yi Ui-xi-68 Vi=yi-68 Ui^2 Vi^2 Uivi


65 67 3 -1 9 1 3
67 68 1 0 1 0 0
66 68 2 0 4 0 0
71 70 3 2 9 4 6
67 64 1 -4 1 16 4
70 67 2 -1 4 1 -2
68 72 0 4 0 16 0
69 70 1 2 1 4 2
Total -1 2 29 39 13

rXY = rUV = (n UV - U. V) /( U (U) V (V) ) )

= 0.3974

10. Write and execute a program in C, C++ Languages or in MATLAB for any problem in
Statistical or Probabilistic Measures.
(Include the problem statement and the programming code with output for the problem).

Solution:
Problem Statement

Write the R programming code for computing the mean, median, mode, variance, standard
deviation, co-efficient of variation, first four moments about the mean and Pearsons co-efficients B1
& B2 for the following frequency distribution, and also comment on the nature of skewness and
kurtosis of the distribution based on the Pearsons co-efficient 1 & 2
16BCE0766

R-PROGRAMMING PART
R Code

#Mean

> h=10

> x=seq(175,245,h)

> f=c(52,68,85,92,100,95,70,28)

> M=(sum(x*f))/sum(f)

>M

[1] 208.9831

> #Median

> h=10

> x=seq(175,245,h)

> f=c(52,68,85,92,100,95,70,28)

> fd=data.frame(x,f)

> cf=cumsum(f)

> n=sum(f)

> mc=min(which(cf>=n/2))

> mcf=f[mc]

> c=cf[mc-1]

> l=x[mc]-h/2

> Md=l+(((n/2)-c)/mcf)*h
16BCE0766
> Md

[1] 209.7826

> #Mode

> h=10

> x=seq(175,245,h)

> f=c(52,68,85,92,100,95,70,28)

> fd=data.frame(x,f)

> mc=which(f==max(f))

> mcf=f[mc]

> f1=f[mc-1]

> f2=f[mc+1]

> l=x[mc]-h/2

> Mo=l+((mcf-f1)/(2*mcf-f1-f2))*h

> Mo

[1] 216.1538

#Variance, Standard Deviation and Pearson's coefficients

> h=10

> x=seq(175,245,h)

> f=c(52,68,85,92,100,95,70,28)

> N=sum(f)

> M=(sum(x*f))/sum(f)

>M

[1] 208.9831

> Var=((1/N)*(sum(x^2*f)))-((1/N)*(sum(x*f)))^2

> Var
16BCE0766
[1] 388.0336

> SD=sqrt(Var)

> SD

[1] 19.69857

> CV=100*(SD/M)

> CV

[1] 9.425917

> #Moments and Pearson's Coefficients

> h=10

> x=seq(175,245,h)

> f=c(52,68,85,92,100,95,70,28)

> N=sum(f)

> M=(sum(x*f))/sum(f)

> M1=sum(((x-M)^1)*f)/N

> M2=sum(((x-M)^2)*f)/N

> M3=sum(((x-M)^3)*f)/N

> M4=sum(((x-M)^4)*f)/N

> B1=(M3^2)/(M2^3)

> B2=(M4)/(M2^2)

> B1

[1] 0.003424753

> B2

[1] 2.034009

You might also like