20MA4T1- STATISTICS AND NUMERICAL METHODS

FOR II - B.E. (MECHANICAL) / IV SEMESTER

SYLLABUS

Course
Semester Programme Course Name L T P C
Code
STATISTICS AND
IV B.E. MECH 20MA4T1 3 1 0 4
NUMERICALMETHODS

COURSE LEARNING OUTCOMES (COs)

RBT Topics
After Successful completion of the course, the students should be able to
Level Covered
Identify and apply various numerical techniques for solving non-linear
CO1 K3 3
equations and systems of linear equations.
Analyse and apply the knowledge of interpolation and determine the
CO2 K4 4
integration and differentiation of the functions by using the numerical data.

Justify the concept of testing of hypothesis for small and large samples and
CO3 K5 1
interpret the results.

Classify the principles of design of experiments and perform analysis of

CO4 K2 2
variance.

Determine the dynamic behaviour of the system through solution of ordinary

CO5 K5 5
differential equations by using numerical methods.

Engineering Mathematics I , Engineering Mathematics II and Transforms and

PRE-REQUISITE
Partial Differential Equations

CO / PO MAPPING (1 – Weak, 2 – Medium, 3 – Strong)

Programme Learning Outcomes (POs) PSOs
COs
PO1 PO2 PO3 PO4 PO5 PO6 PO7 PO8 PO9 PO10 PO11 PO12 PSO1 PSO2
CO1 3 3 3 1 3 3 3
CO2 3 3 3 1 3 3 3
CO3 3 3 3 1 3 3 3
CO4 3 3 3 1 3 3 3
CO5 3 3 3 1 3 3 3

1
 COURSE ASSESSMENT METHODS
DIRECT 1 Continuous Assessment Tests
2 Assignments and tutorials
3 End Semester Examinations
INDIRECT 1 Course Exit Survey

COURSE CONTENT

Topic - 1 TESTING OF HYPOTHESIS 9+3

Sampling distributions – Estimation of parameters – Statistical hypothesis – Large sample tests based on
Normal distribution for single mean and difference of means -Tests based on t, Chi-square and F
distributions for mean, variance and proportion – Contingency table (test for independent) – Goodness of
fit.
Topic - 2 DESIGN OF EXPERIMENTS 9+3
One way and two way classifications – Completely randomized design – Randomized block design –
Latin square design – 22 factorial design.
Topic - 3 SOLUTION OF EQUATIONS AND EIGENVALUE PROBLEMS 9+3
Solution of algebraic and transcendental equations – Fixed point iteration method – Newton Raphson
method – Solution of linear system of equations – Gauss elimination method – Pivoting – Gauss Jordan
method – Iterative methods of Gauss Jacobi and Gauss Seidel – Eigen values of a matrix by Power
method.
Topic - 4 INTERPOLATION, NUMERICAL DIFFERENTIATION AND
9+3
NUMERICAL INTEGRATION
Lagrange’s and Newton’s divided difference interpolations – Newton’s forward and backward difference
interpolation – Approximation of derivates using interpolation polynomials – Numerical single and double
integrations using Trapezoidal and Simpson’s 1/3 rules.
Topic - 5 NUMERICAL SOLUTION OF ORDINARY DIFFERENTIAL EQUATIONS
9+3

Single step methods : Taylor’s series method – Euler’s method – Modified Euler’s method – Fourth order
Runge-Kutta method for solving first order equations – Multi step methods : Milne’s predictor corrector
methods for solving first order equations.
THEORY 45 TUTORIAL 15 PRACTICAL 0 TOTAL 60

BOOK REFERENCES

Gerald. C. F. and Wheatley. P. O., "Applied Numerical Analysis", Pearson Education,Asia, 7th
1
Edition, New Delhi, 2006.

Grewal, B.S., and Grewal, J.S., "Numerical Methods in Engineering and Science", Khanna
2
Publishers, 9th Edition, New Delhi, 2010

3 Burden, R.L and Faires, J.D, “Numerical Analysis”, 9th Edition, Cengage Learning, 2016.

2
4 Vijay K. Rohatgi , EhsanesSaleh,”An Introduction to Probability and Statisics”, 2nd Edition,2009

5 N. G. Das.,”Statistical Methods”, Tata McGraw Hill Publishing Ltd,2008

OTHER REFERENCES
1 https://www.sobtell.com/blog/38-real-life-applications-of-numerical-analysis
https://www.scienceabc.com/eyeopeners/why-do-we-need-numerical-analysis-in-everyday-
2
life.html
3 https://leverageedu.com/blog/application-of-statistics/

3
 UNIT-I
TESTING OF HYPOTHESIS
PART – A

TESTING OF HYPOTHESIS

1. Explain the terms sample size and sampling error

Solution:
Sample size: A finite subset of statistical individuals in a population is called a sample and the number of
individuals in a sample is called sample size.
Sampling error: For the purpose of determining population characteristics, the individuals in the sample are
observed. On examining the sample of a particular stuff we arrive at a decision of purchasing or rejecting
that stuff. The error involved in such approximation is known as sampling error.
2. Define the following terms: Statistics, Parameter and Standard Error
Solution:
Statistic: Measure describing the characteristic of sample
Parameter: Values that describe the characteristic of population
Standard error: Standard deviation of the sampling distribution of a statistic

3. Define Type I and Type II errors in taking a decision.

Solution:
Type I Error: The hypothesis is true but our tests reject it.

Type II Error: The hypothesis is false but our test accepts it.

Decision
Accept 𝐻 Reject 𝐻
𝐻 True Correct decision Type I error
𝐻 False Type II error Correct decision

4. State the proceedure involved in testing of hypothesis.

Solution:
(i) Set up a null hypothesis 𝐻 ,
(ii) Set up the alternative hypothesis 𝐻 ,
(iii) Select the appropriate level of significance (𝛼),
( )
(iv) Compute the test statistic 𝑧 = under 𝐻 ,
( )
(v) We compare the “calculate z” with “critical value 𝑧 ” at given level of significance (𝛼)

[If|𝑧| < 1.96 , 𝐻 may be accepted at 5% level of significance.

If |𝑧| > 1.96 , 𝐻 may be rejected at 5% level of significance.

If |𝑧| > 2.58 , 𝐻 may be accepted at 1% level of significance.
If |𝑧| > 2.58 , 𝐻 may be rejected at 1% level of significance.
5. What is meant by level of significance and critical region?
4
 Solution:
Critical region:
A region corresponding to a statistics t in the sample space S which lead to the rejection of 𝐻 is
called critical Region or Rejection Region.Those region which lead to the acceptance of 𝐻 give as a
region alledAcceptance Region.
Level of significance:
The probability 𝛼 that a random value of the statistic ‘t’ belongs to the critical region is known as
the level of significance. In other words, level of significance is the size of Type I Error. The level of
significance usually employed in testing of hypothesis are 5% and 1%.
6. Define Null and Alternative hypothesis.

Solution:
Null hypothesis is based for analyzing the problem Null hypothesis is the hypothesis of no difference
and is denote by𝐻 .

Any hypothesis which is complementary to the null hypothesis is called an Alternative hypothesis,
denoted by 𝐻

7. A coin is tossed 400 times and it turns up head 216 times. Discuss whether the coin may be unbiased
one at 5% level of significance.
Solution:
Given 𝑛 = 400, 𝑃 =
1 1
𝑄 =1−𝑝=1− =
2 2
X=Number of success=216
(i) The parameter of interest is P.

(ii) 𝐻 : The coin is unbiased

(iii) 𝐻 : The coin is biased

(iv) 𝛼 = 0.05

(v) 𝑍=

(vi) Reject 𝐻 if |𝑧| > 1.96

( )
(vii) Computation:𝑍 = =
( )

5
 (viii) Conclusion: |𝑍| = 1.6 < 1.96
So we accept 𝐻 at 5% level of significance.
Hence coin is unbiased.
8. A standard sample of 200 tins of coconut oil gave an average weight of 4.95Kgs. With a standard
deviation of 0.21 Kg.Do we accept that the net weight is 5 Kgs per tin at 5% level of significance?
Solution:
Given 𝑛 = 200, 𝑥̅ = 4.95 𝑘𝑔, 𝜎 = 0.21 𝑘𝑔, 𝜇 = 5 𝑘𝑔
(i) The parameter of interest is𝜇.
(ii) 𝐻 : 𝜇 = 5 𝑘𝑔𝑠 [The net weight is 5 kgs]
(iii) 𝐻 : 𝜇 ≠ 5 𝑘𝑔𝑠 [Two tailed test]
(iv) 𝛼 = 0.05
̅
(v) 𝑍=
√

(vi) Reject 𝐻 if |𝑧| > 1.96 at 𝛼 = 0.05

.
(vii) Computation: 𝑍 = . = −3.36
√
(viii) Conclusion: |𝑍| = 3.36 > 1.96
So we reject 𝐻 : 𝜇 = 5 𝑘𝑔𝑠 at 5% level of significance.

t-DISTRIBUTION

9. Write down the formula of test statistic t to test the significance of difference between the means of
large samples.
Solution:
|𝑥 − 𝑥 |
𝑍=
+

CHI-SQUARE (𝝌𝟐 ) DISTRIBUTION

10. Define Chi-Square test for goodness of fit.

Solution:
Chi-Square test for goodness of fitis a test to find if the deviation of the experiment from theory is
just by chance or it is due to the inadequacy of the theory to fit the observed data. By this test, we
test whether difference between observed and expected frequencies are significant or not.
Chi-Square test for goodness of fit is defined by
(𝑶 𝑬)𝟐
𝝌𝟐 = ∑ . Where O-Observed frequency
𝑬

E- Expected frequency

6
 11. Write any two applications of 𝐂𝐡𝐢 − 𝐬𝐪𝐮𝐚𝐫𝐞 (𝝌𝟐 )⬚– test.
Solution:
(i) To test the goodness of fit

(ii) To test the “independence of attributes”

(iii) To test the homogeneous of independent estimations.

12. What are the conditions for the validity of 𝚿 𝟐 - test? OR State the conditions for applying 𝚿 𝟐 -
test.

Solution:
(i) The sample observations should be independent.
(ii) Constraints on the cell frequencies, if any, must be linear [e.g., ∑ 𝑂 = ∑ 𝐸 ]
(iii) N, the total frequency, should be at least 50.
(iv) No theoretical cell frequency should be less than 5.

13. What are the expected frequencies of 2*2 contigency table a b

Solution: c d
Expected frequency table:
(𝑎 + 𝑏)(𝑎 + 𝑐) (𝑎 + 𝑏)(𝑏 + 𝑑)
𝑁 𝑁
(𝑎 + 𝑐)(𝑐 + 𝑑) (𝑐 + 𝑑)(𝑏 + 𝑑)
𝑁 𝑁

14. Give the formula for the 𝝌𝟐 - test of independence for a b

c d
Solution:

𝑁(𝑎𝑑 − 𝑏𝑐)
𝜒 = , 𝑁 =𝑎+𝑏+𝑐+𝑑
(𝑎 + 𝑐)(𝑏 + 𝑑)(𝑎 + 𝑏)(𝑐 + 𝑑)

PART – B

TESTING OF HYPOTHESIS FOR SINGLEMEAN

1. Explain clearly the procedure generally followed in testing of a hypothesis.
Solution:
General procedure for hypothesis tests
i. From the problem context, identify the parameter of interest.
ii. State the null hypothesis, 𝐻 .
iii. Specify an appropriate alternative hypothesis, 𝐻 .
iv. Choose a significance level 𝛼.

7
 If |𝑧| < 1.96, 𝐻 may be accepted at 5% level of significance.
If |𝑧| > 1.96, 𝐻 may be rejected at 5% level of significance.
If |𝑧| < 2.58, 𝐻 may be accepted at 1% level of significance.
If |𝑧| > 2.58, 𝐻 may be rejected at 1% level of significance.
v. Determine an appropriate test statistic.
vi. State the rejection for the statistic.
vii. Compute any necessary sample quantities, substitute these into the equation for the test statistic and
compute the value.
viii. Conclusion: Decide whether or not, 𝐻 should be rejected and report that in the problem context.
2. A sample of 900 members has a mean 3.4 cm and standard deviation 2.61 cm. Is the sample from a
large population of mean 3.25 cms and standard deviation of 2.61 cms? (Test at 5% level of
significance. The value of 𝒛 at 5% level is|𝒁∝ | < 1.96).
Solution:
Given 𝑛 = 900𝜇 = 3.25
𝑥̅ = 3.4 𝑐𝑚𝜎 = 2.61
𝑠 = 2.61
(i) The parameter is𝜇 .
(ii) Null hypothesis𝐻 : Assume that the sample has been drawn from the population with mean𝜇 = 3.25.
(iii) Alternative hypothesis 𝐻 : 𝜇 ≠ 3.25
(iv) Level of significance ∝= 0.05.
̅
(v) The test statistic is 𝑧 =
√
(vi) Reject if |𝑧| > 1.96 at 5% level of siginificance.
(vii) Computation:
𝑥̅ − 𝜇 3.4 − 3.25
𝑧= ⬚ = .
√
√

= 1.724

𝑧 = 1.724 < 1.96

(viii) Conclusion
Here |𝑧| = 1.724 < 1.96, so we accept the null hypothesis 𝐻 at 5% level of significance.
TESTING OF HYPOTHESIS FOR DIFFERENCE OF MEAN

3. The means of two large samples of 1000 and 2000 members are 67.5 and 68.0 inches respectively. Can
the samples be regarded as drawn from the same populations of standard deviation 2.5 inches?
Solution:
Given sample sizes𝑛 = 1000, 𝑛 = 2000
𝑥̅ = 67.5, 𝑥̅ = 68.0
𝑠 = 𝜎 = 2.5 𝑠 = 𝜎 = 2.5
(i) The parameter is𝜇 &𝜇 .
(ii) Null hypothesis𝐻 : 𝜇 = 𝜇 (there is no significant difference).
(iii) Alternative hypothesis 𝐻 : 𝜇 ≠ 𝜇
(iv) Level of siginificance∝= 0.05.
8
 ̅ ̅
(v) The test statistic is 𝑧 =

(vi) Reject if |𝑧| > 1.96 at 5% level of siginificance.

(vii) Computation:
𝑥̅ − 𝑥̅
𝑧=
+

67.5 − 68.0 −0.5

= = = −5.16
. . 0.09685
+

|𝑧| = 5.16
(viii) Conclusion
Here |𝑧| = 5.16 > 1.96, so we reject the null hypothesis 𝐻 at 5% level of significance.

4. A random sample of 100 bulbs from a company P shows a mean life 1300 hours and standard deviation of 82
hours. Another random sample of 100 bulbs from company Q showed a mean life 1248 hours and standard
deviation of 93 hours. Are the bulbs of company P superior to bulbs of company Q at 5% level of
significance?

Solution:

𝐻 :𝜇 = 𝜇 ,

𝐻 : 𝜇 > 𝜇 , (right tailed test)

L.O.S 𝛼 = = 0.05.

̅ ̅
Test statistics 𝑧 =

Where 𝑛 = 100, 𝑛 = 100

𝑥̅ = 1300, 𝑥̅ = 1248
𝑠 = 𝜎 = 82 𝑠 = 𝜎 = 93
𝑥̅ − 𝑥̅
𝑧=
+

1300 − 1248
=
+

52
= = 4.1939
12.39879
9
 Table value of z at 5% LOS is 1.675.
Since 4.19 > 1.645, 𝐻 is rejected and the bulbs of company A is superior to the bulbs of company B.

𝑥̅ − 𝑥̅
𝑧=
+

67.5 − 68.0 −0.5

= = = −5.16
. . 0.09685
+

5. A mathematics test was given to 50 girls and 75 boys. The girls made an average grade of 76 with a
SD of 6, while boys made an average grade of 82 with a SD of 2. Test whether there is any significant
difference between the performance of boys and girls.
Solution: (AU N/D 2012, M/ J 2016)
No of cases Mean S.D
Sample I 50 76 6
Sample II 75 82 2

Given sample sizes 𝑛 = 50, 𝑛 = 75

𝑥̅ = 76, 𝑥̅ = 82
𝑠 =𝜎 =6 𝑠 =𝜎 =2
The parameter is𝜇 &𝜇 .

(i) Null hypothesis𝐻 : 𝜇 = 𝜇 (there is no significant difference).

(ii) Alternative hypothesis 𝐻 : 𝜇 ≠ 𝜇

(iii) Level of significance∝= 0.05.

̅ ̅
(iv) The test statistic is 𝑧 =

(v) Reject if |𝑧| > 1.96 at 5% level of siginificance.

(vi) Computation:

10
 𝑥̅ − 𝑥̅ 76 − 82
𝑧= =
+ +

−6 6
= =−
0.88
+

= −6.82

|𝑧| = 6.82
(vii) Conclusion

Here |𝑧| = 6.82 > 1.96, so we reject the null hypothesis 𝐻 at 5% level of significance.

6. Test if the variances are significantly different for

X1 24 27 26 21 25
X2 27 30 32 36 28 23
Solution:

x 𝒙−𝒙 (𝒙 − 𝒙)𝟐 Y 𝒚−𝒚 (𝒚 − 𝒚)𝟐

24 -0.6 0.36 27 -2.3 5.29
27 2.4 5.76 30 0.7 0.49
26 1.4 1.96 32 -2.7 7.29
21 -3.6 12.96 36 6.7 44.89
25 0.4 0.16 28 -1.3 1.69
23 -6.3 39.69
123 0 21.2 176 -5.2 99.34

𝑥= = 24.6, 𝑦 = = 29.3

(𝑥 − 𝑥) = 21.2, (𝑦 − 𝑦) = 99.34
1
𝑠 = (𝑥 − 𝑥) + (𝑦 − 𝑦)
𝑛 +𝑛 −2
1
= [21.2 + 99.34] = 13.39
5+6−2
𝑠 = 3.66
(i) The parameter is𝜇 &𝜇 .
(ii) Null hypothesis𝐻 : 𝜇 = 𝜇 (there is no significant difference).
11
 (iii) Alternative hypothesis 𝐻 : 𝜇 ≠ 𝜇
(iv) Level of significance ∝= 0.05.
(v) The test statistic is 𝑧 =

(vi) Reject if |𝑧| > 2.262 at 5% level of significance for 9 degrees of freedom
(vii) Computation:
𝑥−𝑦 24.6 − 29.3 −4.7
𝑧= = = = −2.1197
2.2172
𝑠 + 3.66 +
|𝑧| = 2.12
(viii) Conclusion
Here|𝑧| = 2.12 < 2.262, so we accept the null hypothesis 𝐻 at 5% level of significance
Hence there is no significant difference.

TESTING OF HYPOTHESIS FOR DIFFERENCE OF PROPORTION

t-DISTRIBUTION( n<30 )
7. Test if the difference in the means is significant for the following data:
Sample I 76 68 70 43 94 68 33
Sample II 40 48 92 85 70 76 68 22
Solution:
Calculation for sample means and S.D.’s
𝑥 𝑥 − 𝑥̅ (𝑥 − 𝑥̅ ) 𝑦 𝑦−𝑦 (𝑦 − 𝑦)
76 11.4 129.96 40 -22.6 510.76
68 3.4 11.56 48 -14.6 213.16
70 5.4 29.16 92 29.4 864.36
43 -21.6 466.56 85 22.4 501.76
94 29.4 864.36 70 7.4 54.76
68 3.4 11.56 76 13.4 179.56
33 -31.6 998.56 68 5.4 29.16
22 -40.6 1648.36
452 2511.72 501 4001.88
Given 𝑛 = 7, 𝑛 = 8 (< 30 𝑠𝑜 𝑤𝑒 𝑢𝑠𝑒 𝑡 − 𝑡𝑒𝑠𝑡)
Mean 𝑥̅ = = 64.6 𝑦= = 62.6
(𝑥 − 𝑥̅ ) = 2511.72 (𝑦 − 𝑦) = 4001.88

12
 ∑(𝑥 − 𝑥̅ ) + ∑(𝑦 − 𝑦) 2511.72 + 4001.88
𝑠 = = = 501.04
𝑛 +𝑛 −2 7+8−2
𝑠 = 22.38
(i) The parameter is𝜇 &𝜇 .
(ii) Null hypothesis𝐻 : 𝜇 = 𝜇 (there is no significant difference in the variability in yields).
(iii) Alternative hypothesis 𝐻 : 𝜇 ≠ 𝜇
(iv) Level of significance ∝= 0.05. d.f= 7 + 8 − 2 = 13
̅ ̅
(v) The test statistic is 𝑡 =

(vi) Reject if |𝑡| > 2.16 at 5% level of significance.

(vii) Computation:
𝑥̅ − 𝑥̅ 64.6 − 62.6
𝑡= = = 0.1727
𝑠 + 22.38 +
|𝑡| = 0.1727
(viii) Conclusion
Here |𝑡| = 0.1727 < 2.16, we reject the null hypothesis 𝐻 at 5% level of significance.
There is no significant difference between the two means.

CHI-SQUARE (𝝌𝟐 ) DISTRIBUTION

8. Using the data given in the following table to test at 1% level significance whether a person’s ability in
Mathematics is independent of his/her interest in Statistics.
Ability in
Mathematics
Low Average High
Interest in Low 63 42 15
Statistics
Average 58 61 31
High 14 47 29

Solution:
Table of expected frequencies.
Low Average High Total

13
 Low 63 42 15 120
Average 58 61 31 150
High 14 47 29 90
Total 135 150 75 360

135 × 120 150 × 120 75 × 120 120

360 360 360
= 45 = 50 = 25

135 × 150 150 × 150 75 × 150 150

360 360 360
= 56.25 = 62.5 = 31.25

135 × 90 150 × 90 75 × 90 90
360 360 360
= 33.75 = 37.5 = 18.75

135 150 75 360

Calculated Ψ

Observed Expected (𝑶 − 𝑬)𝟐

(O-E)
frequency(O) frequency(E) 𝑬

63 45 324 7.5

42 50 64 1.28

15 25 100 4.00

58 56.25 3.0625 0.05

61 62.5 2.25 0.04

14
 31 31.25 0.0625 0.002

14 33.75 390.0625 11.56

47 37.5 90.25 2.41

29 18.75 105.0625 5.60

32.142
(𝑶 𝑬)𝟐
Now Ψ = ∑ = 𝟑𝟐. 𝟏𝟒𝟐
𝑬

Ψ = 32.14

Tab Ψ fort d.f at 5% level is 9.488

Since Cal Ψ > 𝑇𝑎𝑏 Ψ ∴ 𝑤𝑒 𝑟𝑒𝑗𝑒𝑐𝑡 𝐻 .

Hence ability in Mathematics and interest in statistics are depended.

9. Fit a binomial distribution for the following data and also test the goodness of fit.
X: 0 1 2 3 4 5 6 Total
F(X): 5 18 28 12 7 6 4 80
Solution:
Given n=2
(i) The parameter is 𝜒
(ii) 𝐻 :Binomial is good fit
(iii) 𝐻 :Binomial is not a good fit
(iv) Level of significance ∝= 0.0
( )
(v) The test statistic is 𝜒 = ∑

(vi) Reject 𝐻 if 𝜒 > 𝜒 .

(from 𝜒 table)
(vii) Computation :
𝑝(𝑥) = 𝑛𝐶 𝑝 𝑞
80
𝐸= = 11
7
𝑛𝑝 = 2.4
15
 6𝑝 = 2.4
𝑝 = 0.4, 𝑞 = 0.6
Observed Expected (𝑂 − 𝐸)
frequency(O) frequency(E) 𝐸

5 11 3.27
18 11 4.46
28 11 26.27
12 11 0.99
7 11 1.46
6 11 2.27
4 11 4.46

(𝑂 − 𝐸)
𝜒 = = 6.17
𝐸
𝑛=2
𝜒 .
= 5.98

if𝜒 > 𝜒 .
Reject 𝐻
10. Theory predicts that the proportion of beans in four groups A, B, C, D should be 9:3:3:1. In an
experiment among 1600 beans, the numbers in the four groups were 882, 313, 287 and 118. Does the
experiment support the theory?

Solution:

Given A, B, C, D in the ration 9: 3: 3: 1

Total

𝐸: 900 300 300 100 1600

𝑂: 882 313 287 118 1600

(𝑂 − 𝐸)
𝜒 = ⬚
𝐸

18 13 13 18
= + + + = 4.73
900 300 300 100
16
 𝐸 = 𝑂 , 𝑑. 𝑓 = 4 − 1 = 3

𝐻 : The experiment supports the theory

𝐶𝑎𝑙 𝜒 = 4.73

Table 𝜒 3 d.f=7.82

𝐶𝑎𝑙 𝜒 < Table 𝜒

So we accept 𝐻

F-TEST
11. A group of 10 rats fed on diet A and another group of 8 rats fed on diet B, recorded the following
increase in weight (gms)
Diet A: 5 6 8 1 12 4 3 9 6 10
Diet B: 2 3 6 8 10 1 2 8
Does it show superiority of diet A over diet B?
Solution:
Given 𝑛 = 10, 𝑛 = 8
𝑥 𝑥 𝑥 𝑥
5 25 2 4
6 36 3 9
8 64 6 36
1 1 8 64
12 144 10 100
4 16 1 1
3 9 2 4
9 81 8 64
6 36
10 100
64 512 40 282

17
 ∑𝑥 64 ∑𝑥 40
𝑥̅ = = = 6.4 𝑥̅ = = =5
𝑛 10 𝑛 8
∑𝑥 512
𝑆 = − (𝑥 ) = − (6.4) = 10.24
𝑛 10
∑𝑥 282
𝑆 = − (𝑥 ) = − 25 = 10.25
𝑛 8
𝑆 <𝑆
(i) The parameter is𝜎 &𝜎 .
(ii) Null hypothesis𝐻 : 𝜎 = 𝜎 (there is no significant difference).
(iii) Alternative hypothesis𝐻 : 𝜎 ≠𝜎
(iv) Level of significance ∝= 0.05. Degree of freedom(𝑣 ) = 9, 𝑑. 𝑓 (𝑣 ) = 7 𝑖𝑒. , 𝐹( , ) = 3.29
(v) Accept 𝐻 if calculated F <𝐹( , ) = 3.29 (from table ‘F’ )
.
(vi) Computation:The test statistic is 𝐹 = = = 1.0009
.

(vii) Conclusion
Here |𝐹| = 1.0009 < 3.29, so we accept the null hypothesis 𝐻 .
We conclude that the two samples have come from populations with equal variances.
12. Two independent samples of sizes 9 and 7 from a normal population had the following values of the
variables.
Sample I 18 13 12 15 12 14 16 14 15
Sample II 16 19 13 16 18 13 15
Do the estimates of the population variance differ significantly at 5% level ?

Solution:
Given 𝑛 = 9, 𝑛 = 7
𝑥 𝑥 𝑥 𝑥
18 324 16 256
13 169 19 361
12 144 13 169
15 225 16 256
12 144 18 324
14 196 13 169
16 256 15 225
14 196
15 225
129 1879 110 1760

18
 ∑𝑥 129 ∑𝑥 110
𝑥̅ = = = 14.3333𝑥̅ = = = 15.7143
𝑛 9 𝑛 7
∑𝑥 1879
𝑆 = − (𝑥 ) = − (14.3333) = 3.3342
𝑛 9
∑𝑥 1760
𝑆 = − (𝑥 ) = − (15.7143) = 4.4894
𝑛 7
𝑆 <𝑆
(i) The parameter is𝜎 &𝜎 .

(ii) Null hypothesis𝐻 : 𝜎 =𝜎 (there is no significant difference).

(iii) Alternative hypothesis𝐻 : 𝜎 ≠𝜎

(iv) Level of significance ∝= 0.05. Degree of freedom(𝑣 ) = 8, 𝑑. 𝑓 (𝑣 ) = 7 𝑖𝑒. , 𝐹( , ) = 3.58

(v) Accept 𝐻 if calculated F <𝐹( , ) = 3.58 (from table ‘F’ )

.
(vi) Computation:The test statistic is 𝐹 = = = 1.3464
.

(vii) Conclusion

Here |𝐹| = 1.3464 < 3.58, so we accept the null hypothesis 𝐻 .

We conclude that the difference is not significant.

13. Two random samples gave the following results:

Samples Size Sample Sum of squares of deviation
mean from the mean
1 10 15 90
2 12 14 108
Examine whether the samples come from the same normal population.
Solution:
A normal population has two parameters namely the mean 𝜇 and the variance 𝜎 .If we want to test
samples from the same normal population, we have to test
(i) the equality of population variances (using F-test)
(ii) the equality of population means (using t-test)
Since t-test assumes 𝜎 =𝜎 we shall first apply F-test and then t-test.
(i) F-test

Given𝑛 = 10, 𝑛 = 12, 𝑥 = 15, 𝑥 = 14

19
 (𝑥 − 𝑥̅ ) 90
S = = = 10
𝑛 −1 9

(𝑥 − 𝑥̅ ) 108
S = = = 9.8181
𝑛 −1 11

S >S

(i) The parameter of interest is 𝜎 and 𝜎

(ii) 𝐻 :𝜎 = 𝜎
(iii) 𝐻 :𝜎 ≠ 𝜎
(iv) 𝛼 = 0.05, 𝑑. 𝑓 (𝑣 ) = 𝑛 − 1 = 9 𝑑. 𝑓 (𝑣 ) = 𝑛 − 1 = 11
(v) The test statistic is 𝐹 =
(vi) Reject 𝐻 if F > 2.90 (from table ‘F’)
(vii) Computation:𝐹 = = 1.019
.
(viii) Conclusion: Here F=1.019< 2.90, so we accept 𝐻 at 5% level of significance.
[Note: If F-test failed, then t-test should not be used]
(ii) t-test

Given 𝑛 = 10, 𝑛 = 12, S = 10, S = 9.8181

𝑛 𝑠 +𝑛 𝑠 100 + 117.82
𝑆 = = = 10.9
𝑛 +𝑛 −2 10 + 12 − 2

(i) The parameter is𝜇 &𝜇 .

(ii) Null hypothesis𝐻 : 𝜇 = 𝜇 (there is no significant difference).
(iii) Alternative hypothesis 𝐻 : 𝜇 ≠ 𝜇
(iv) Level of significance ∝= 0.05. Degree of freedom=𝑛 + 𝑛 − 2 = 20
̅ ̅
(v) The test statistic is 𝑧 =

(vi) Reject if |𝑡| > 2.086[from table ‘t’ we get t=2.086] at 5% level of significance.
(vii) Computation:
𝑥̅ − 𝑥̅ 1
𝑡= =
1.3472
𝑆 +

= 0.707
(viii) Conclusion
Here |𝑡| = 0.707 < 2.086, so we accept the null hypothesis 𝐻 at 5% level of significance.
Hence the difference is not significant.

20
 Final conclusion: From the above two test, we can conclude that the two sample drawn from the
same normal population.

UNIT-II

DESIGN OF EXPERIMENTS

PART – A

1. Define Analysis of variance (ANOVA)

Solution:

Analysis of variance (ANOVA) is a technique that will enable us to test for the significance of the

difference among more than two sample means.

2. State the basic principles of design of experiments.

Solution:

There are three basic principles of experimental design. They are (i) Randomization, (ii) Replication,

(iii) Local control (error control)

3. What is the aim of the design of experiments?

Solution:
21
 The main aim of the design of experiments is to control the extraneous variables and hence to

minimize the experimental error so that the result of the experiments could be attributed only to the

experimental variable.

4. State the assumptions involved in ANOVA.

Solution:

For the validity of the F-test in ANOVA, the following assumptions are made:

(i) The observations are independent.

(ii) Parent population from which observations are taken in normal and

(iii) Various treatment and environmental effects are additive in nature.

5. What are the uses of ANOVA?

Solution :

Analysis of variance is useful , for example , for determining

(I )Which of various training methods produces the fastest learning record.

(ii) Whether the effects of some fertiliserson the yields are significantly different,

(iii) Whether the mean qualities of outputs of various machines differ significantly etc. In fact this

technique finds application in nearly every type of experimental design in natural sciences as well

as in social sciences.

COMPLETELY RANDOMIZED DESIGN

6. Compare one-way classification model with two-way classification model.

Solution:

22
 One way Two way

1. We cannot test two sets of hypothesis 1. Two sets of hypothesis can be tested.

2. Data are classified according to one 2. Data are classified according to the

factor different factor.

7. What is a completely randomized design.

Solution:

In a completely randomized design the treatments are given to the experimental units

by a procedure of random allocation. It is used when the units are homogeneous.

8. State any two advantages of a Completely Randomized Experimental Design.

Solution:

(i) It is easy to lay out the design.

(ii) It allows for complete flexibility. Any number of factor classes and replications may be used.

(iii) The statistical analysis is relatively simple, even if we do not have the same number of

replicates for each factor class or if the experimental errors are not the same from class to

class of this factor.

(iv) The method of analysis remains when data are missing or rejected and the loss of

information due to missing data is smaller than any other design.

9. Define 𝟐𝟐 factorial design .

Solution:

A two-factor factorial design is an experimental design in which data is collected for all possible

combinations of the levels of the two factors of interest.

10. Why 𝟐 × 𝟐 Latin square is not possible? Explain.

23
 Solution:

In Latin square, the formula for degrees of freedom for residual (SSE) is d.f=(𝑛 − 1)(𝑛 − 2)

Substituting 𝑛 = 2, 𝑑. 𝑓 = 0

𝑀𝑆𝐸 = ∞

∴ 2 × 2 Latin square is not possible

PART – B

COMPLETELY RANDOMIZED DESIGN (C.R.D) (OR) [ONE WAY CLASSIFICATION]

1. The following table shows the lives in hours of four brands of electric lamps

Brand A 1610 1610 1650 1680 1700 1720 1800

B 1580 1640 1640 1700 1750

C 1460 1550 1600 1620 1640 1660 1740 1820

D 1510 1520 1530 1570 1600 1680

Perform an analysis of variance test the homogeneity of the mean lives of the four

brands of lamps.

Solution:

𝐻 : There is no significant difference between the four brands

𝐻 : There is significant difference between the four brands

Subtract 1600 and then divided by 10 we get

𝑿𝟏 𝑿𝟐 𝑿𝟑 𝑿𝟒 Total 𝑿𝟏 𝟐 𝑿𝟐 𝟐 𝑿𝟑 𝟐 𝑿𝟒 𝟐

A B C D

24
1 -2 -14 -9 -24 1 4 196 81

1 4 -5 -8 -8 1 16 25 64

5 4 0 -7 2 25 16 0 49

8 10 2 -3 17 64 100 4 9

10 15 4 0 29 100 225 16 0

12 - 6 8 26 144 - 36 64

20 - 14 - 34 400 - 196 -

- - 22 - 22 - - 484 -

57 31 29 -19 98 735 361 957 267

Step 1: 𝑁 = 26

Step 2: T=98

𝑇 9604
= = 369.39
𝑁 26

TSS=∑ 𝑋 +∑𝑋 +∑𝑋 +∑𝑋 −

= 753 + 361 + 957 + 267 − 369.39

= 1950.61

(∑ ) (∑ ) (∑ ) (∑ )
SSC= + + + − 𝑁 →Number of elements in each column

(57) (31) (29) (−19)

= + + + − 369.39
7 5 6 6

= 452.25

𝑆𝑆𝐸 = 𝑇𝑆𝑆 − 𝑆𝑆𝐶

= 1950.61 − 452.25 = 1498.36

Step 6: ANOVA

25
 Source of Sum of squares Degree of Mean squares Variance- Table

variaions freedom ratio value at

5% level

Between SSC=452.25 C-1=4-1=3 MSC= = 𝑀𝑆𝐶 𝐹 (3,22)

𝐹 =
𝑀𝑆𝐸
Columns . = 3.05
= 150.75 150.75
=
68.11
Error SSE=1498.36 N-C=26-4=22 MSE= =
= 2.21
.
= 68.11 Since <

Step 7: Conclusion: Cal 𝐹 < Tab 𝐹

∴ So we accept 𝐻

RANDOMIZED BLOCK DESIGN (R.B.D) (or) [TWO WAY CLASSIFICATION]

2. A set of data involving four “four tropical feed stuffs A, B, C, D” tried on 20 chicks is given

below. All the twenty chicks are treated alike in all respect expect the feeding treatments

and each feeding treatment is given to 5 chicks. Analyze the data.

Weight gain of baby chicks fed on different feeding materials composed of tropical feed

stuffs.

Total Ti

A 55 49 42 21 52 219

B 61 112 30 89 63 355

C 42 97 81 95 92 407

26
 D 169 137 169 85 154 714

Grand Total G=1695

Solution :

Null hypothesis𝐻 :

(i) 𝐻 : There is no significant difference between treatments(columns)

(ii) 𝐻 : There is no significant difference between stuffs.(rows)

Code the data by subtracting 50 from each value.

Stuffs Treatments

𝑿𝟏 𝑿𝟐 𝑿𝟑 𝑿𝟒 𝑿𝟓 Total 𝑿𝟏 𝟐 𝑿𝟐 𝟐 𝑿𝟑 𝟐 𝑿𝟒 𝟐 𝑿𝟓 𝟐

(𝑌 ) =A 5 -1 -8 -29 2 -31 25 1 64 841 4

(𝑌 ) =B 11 62 -20 39 13 105 121 3844 400 1521 169

(𝑌 ) =C -8 47 31 45 42 157 64 2209 961 2025 1764

(𝑌 ) =D 119 87 119 35 104 464 14161 7569 14161 1225 10816

Total 127 195 122 90 161 695(T) 14371 13623 15586 5612 12753

Step: 1

N=20.

Step: 2

T =695

Step: 3

𝑇 (695)
= = 24151.25
𝑁 20

Step: 4

TSS=∑ 𝑋 +∑𝑋 +∑𝑋 +∑𝑋 + ∑𝑋 −

= 14371 + 13623 + 15586 + 5612 + 12753 − 24151.25

= 37793.75

Step:5

27
 (∑ ) (∑ ) (∑ ) (∑ ) (∑ )
SSC= + + + + − ; 𝑁 →Number of elements in each column

(127) (195) (122) (90) (161)

= + + + + − 24151.25
4 4 4 4 4

= 4032.25 + 9506.25 + 3721 + 2025 + 6480.25 − 24151.25

= 1613.50

Step: 6

(∑ ) (∑ ) (∑ ) (∑ )
SSR= + + + − 𝑁 →Number of elements in each row

(−31) (105) (157) (464)

= + + + − 24151.25
5 5 5 5

= 192.2 + 2205 + 4929.8 + 43059.2 − 24151.25

= 26234.95

SSE= 𝑻𝑺𝑺 − 𝑺𝑺𝑪 − 𝑺𝑺𝑹

= 37793.75 − 1613.50 − 26234.95

= 9945.3

Step: 7 ANOVA TABLE

Source of Sum of squares Degree of Mean squares F-ratio Table value at

variaions freedom 5% level

Between SSC=1613.50 c-1=5-1=4 MSC= = 𝑀𝑆𝐸 𝐹 (12,4) = 5.91

. 𝐹 =
𝑀𝑆𝐶
Block(Columns)
403.375 = 2.055

Between SSR=26234.95 r-1=4-1=3 MSR= = 𝑀𝑆𝑅 𝐹 (3,12) = 3.49

. 𝐹 =
𝑀𝑆𝐸
Varieties(Rows) .
= 10.55

Residual SSE=9945.3 (N-c- MSE= =

.
= -

r+1)=12 828.775

Step: 8 Conclusion:

Cal 𝐹 < 𝑇𝑎𝑏𝑙𝑒 𝐹 so we accept 𝐻

28
 Cal 𝐹 > 𝑇𝑎𝑏𝑙𝑒 𝐹 so we reject 𝐻

3. Carry out ANOVA (Analysis of variance) for the following

A B C D

1 44 38 47 36

2 46 40 52 43

Workers 3 34 36 44 32

4 43 38 46 33

5 38 42 49 39

Solution:

Null hypothesis 𝐻 :

(i) The mean productivity is the same for four different machines

(ii) The 5 men do not differ with respect to mean productivity.

Code the data by subtracting 40 from each value.

The coded data is

Machine Type Total

Workers 𝑋 𝑋 𝑋 𝑋 𝑋 𝑋 𝑋 𝑋

𝑌 4 -2 7 -4 5 16 4 49 16

𝑌 6 0 12 3 21 36 0 144 9

𝑌 -6 -4 4 -8 -14 36 16 16 64

𝑌 3 -2 6 -7 0 9 4 36 49

𝑌 -2 2 9 -1 8 4 4 81 1

Total 5 -6 38 -17 T=20 101 28 326 139

Step: 1

N =20

Step: 2

T=20
29
 Step: 3

Correction factor (C.F)= = = 20

Step: 4

TSS=∑ 𝑋 +∑𝑋 +∑𝑋 +∑𝑋 −

= 101 + 28 + 326 + 139 − 20

= 574

(∑ ) (∑ ) (∑ ) (∑ ) (∑ )
SSC= + + + + − 𝑁 →Number of elements in each column

(5) (−6) (38) (−17)

= + + + − 𝐶. 𝐹
4 4 4 4

= 5 + 7.2 + 288.8 + 57.8 − 20

= 338.8

(∑ ) (∑ ) (∑ ) (∑ )
SSR= + + + − 𝑁 →Number of elements in each row

(5) (21) (−14) (0) (8)

= + + + + − 𝐶. 𝐹
4 4 4 4 4

= 6.25 + 110.25 + 49 + 16 − 20

= 161.5

𝑺𝑺𝑬 = 𝑻𝑺𝑺 − 𝑺𝑺𝑪 − 𝑺𝑺𝑹

= 574 − 338.8 − 161.5

= 73.7

ANOVA TABLE

Source of Sum of Degree of Mean squares F-ratio Table

variations squares freedom(D.f) value at

5% level

Between SSC=338.8 𝐶 − 1=4-1=3 MSC= = 𝑀𝑆𝐶 𝐹 (3,12)

. 𝐹 =
𝑀𝑆𝐸
Block(Columns) = 3.49
= 18.38
30
 .
= 112.933

Between SSR=161.5 𝑅 − 1=5-1=4 MSR= = 𝑀𝑆𝑅 𝐹 (4,12)

. 𝐹 =
𝑀𝑆𝐸
Varieties(Rows) . = 3.26
= 40.375 = 6.574

Residual SSE=73.7 (𝐶 − 1)( 𝑅 − 1)=12 MSE= = -

.

.
= 6.142

Step: 8 Conclusion

(i) Table 𝐹 (3,12) 𝑎𝑡 5% 𝑙𝑒𝑣𝑒𝑙 = 3.49

Calculated value 𝐹 = 18.38 > 𝐹 = 3.49 Reject 𝐻

∴ Mean productivity is not the same for the four different types of machines.

(ii) Table 𝐹 (4,12) 𝑎𝑡 5% 𝑙𝑒𝑣𝑒𝑙 = 3.26

Calculated value 𝐹 = 6.58 >𝐹 = 3.26 Reject 𝐻

∴ The workers differ with respect to mean productivity.

4. A company appoints 4 salesmen A, B, C and D and observes their sales in 3 seasons,

summer, winter and monsoon. The figures are given in the following table:

season Salesmen

A B C D

Summer 45 40 28 37

Winter 43 41 45 38

Monsoon 39 39 43 41

Carry out an analysis of variances.

Solution:

Null hypothesis 𝐻 : (i)The salesman do not differ significantly in their performance.

(𝑖𝑖)There is no significant difference between the seasons.

Code the data by subtraction 43 from each value to simplify calculations.

31
Seasons Salesmen
𝑋 𝑋 𝑋 𝑋
𝑋 𝑋 𝑋 𝑋 Total

Summer -2 -3 15 6 16 4 9 225 36

Winter 0 2 -2 5 5 0 4 4 25

Monsoon 4 4 0 2 10 16 16 0 4

Total 2 3 13 13 31 20 29 229 65

Step 1: 𝑁 = 12

Step 2: T=31

( )
Step 3: = = 80

Step 4:

TSS=∑ 𝑋 +∑𝑋 +∑𝑋 +∑𝑋 −

= 20 + 29 + 229 + 65 − 80

= 263

Step 5:

(∑ ) (∑ ) (∑ ) (∑ )
SSC= + + + − 𝑁 →Number of elements in each column

(2) (3) (13) (13)

= + + + − 80
3 3 3 3

= 37.6

Step 6:

(∑ ) (∑ ) (∑ )
SSR= + + − 𝑁 →Number of elements in each row

(16) (5) (10)

= + + − 80
4 4 4

= 15.25

𝑆𝑆𝐸 = 𝑇𝑆𝑆 − 𝑆𝑆𝐶 − 𝑆𝑆𝑅

32
 = 263 − 37.6 − 15.25 = 210.15

Step 7: ANOVA TABLE

Source of Sum of Degree of Mean squares Variance- Table value at

variations squares freedom ratio 5% level

Between SSC=37.67 c-1=4-1=3 MSC= 𝑀𝑆𝐸 𝐹 (6,3) = 8.94

𝐹 =
𝑀𝑆𝐶
Block(Columns) 37.67
= = 12.56 = 2.79
3

Between SSR=15.25 r-1=3-1=2 MSR= 𝑀𝑆𝐸 𝐹 (6,2) = 19.33

𝐹 =
𝑀𝑆𝑅
Varieties(Rows) 15.25
= = 7.63 = 4.587
2

Residual SSE=210.15 (𝐶 − 1)( 𝑅 − 1)=6 MSE= -

210.15
= = 35
6

Step 8: Conclusion:

Cal 𝐹 < Tab 𝐹

∴ So we accept 𝐻

5. Four varieties A, B, C, D of a fertilizer are tested in a randomized block design with 4

replication. The plot yields in pounds are as follows:

Column Row 1 2 3 4

1 A(12) D(20) C(16) B(10)

2 D(18) A(14) B(11) C(14)

3 B(12) C(15) D(19) A(13)

4 C(16) B(11) A(15) D(20)

Analyze the experimental yield.

Solution:

33
Null hypothesis 𝐻 : Four varieties are similar

Alternative hypothesis 𝐻 : Four varieties are not similar.

Calculation of correction factor

Block Total of

Variety 1 2 3 4 varieties 𝑋 𝑋 𝑋 𝑋

(𝑋 ) (𝑋 ) (𝑋 ) (𝑋 )

A 12 14 15 13 54 144 196 225 169

B 12 11 11 10 44 144 121 121 100

C 16 15 16 14 61 256 225 256 196

D 18 20 19 20 77 324 400 361 400

Total 58 60 61 57 236 868 942 963 865

Step 1: 𝑁 = 16

Step 2: T=236

( )
Step 3: = = 3481

Step 4:

TSS=∑ 𝑋 +∑𝑋 +∑𝑋 +∑𝑋 −

= 868 + 942 + 963 + 865 − 3481

= 157

Step 5:

(∑ ) (∑ ) (∑ ) (∑ )
SSC= + + + − 𝑁 →Number of elements in each column

(58) (60) (61) (57)

= + + + − 3481
4 4 4 4

= 841 + 900 + 930 + 812 − 3481 = 2

Step 6:

(∑ ) (∑ ) (∑ ) (∑ )
SSR= + + + − 𝑁 →Number of elements in each row

34
 (54) (44) (61) (77)
= + + + − 3481
4 4 4 4

= 729 + 484 + 930.25 + 1482.25 − 3481

= 144.5

𝑺𝑺𝑬 = 𝑻𝑺𝑺 − 𝑺𝑺𝑪 − 𝑺𝑺𝑹

= 157 − 2 − 144.5 = 10.5

Step 7: ANOVA TABLE

Source of Sum of Degree of Mean squares Variance-ratio Table value

variations squares freedom at 5% level

Between 𝑆𝑆𝐶 = 144.5 𝐶−1=3 𝑆𝑆𝐶 144.5 𝑀𝑆𝐶 𝐹 (3,9)

𝑀𝑆𝐶 = = 𝐹 =
𝐶−1 3 𝑀𝑆𝐸
Varieties = 3.86
= 48.17 48.17
= = 39.48
1.22

Between 𝑆𝑆𝑅 = 2 𝑅−1=3 𝑆𝑆𝑅 2 𝑀𝑆𝐸 1.22 𝐹 (9,3)

𝑀𝑆𝑅 = = 𝐹 = =
𝑟−1 3 𝑀𝑆𝑅 0.67
Blocks = 8.81
= 0.67 = 1.82

Residual 𝑆𝑆𝐸 = 11 (𝐶 − 1)( 𝑅 − 1) 𝑀𝑆𝐸

=9 𝑆𝑆𝐸
=
(𝐶 − 1)( 𝑅 − 1)
-
11
= = 1.22
9

Step 8: Conclusion:

Cal 𝐹 > Tab 𝐹 ; Cal 𝐹 < Tab 𝐹

∴ So we reject 𝐻

Hence four varieties are not similar. But the varieties are similar along block wise.

LATIN SQUARE DESIGNS

6. Analyze the variance in the Latin square of yields (in kgs) of paddy where P,Q,R,S

denote the different methods of cultivation:

S122 P121 R123 Q122

35
 Q124 R123 P122 S125

P120 Q119 S120 R121

R122 S123 Q121 P122

Examine whether different method of cultivation have significantly different yields.

Solution:

Null hypothesis 𝐻 : There is no significant difference between rows, between columns and

treatments.

Let us take 120 as origin for simplifying the calculations.

Table I

(𝑋 ) (𝑋 ) (𝑋 ) (𝑋 ) Total 𝑋 𝑋 𝑋 𝑋

𝑌 2 1 3 2 8 4 1 9 4

𝑌 4 3 2 5 14 16 9 4 25

𝑌 0 -1 0 1 0 0 1 0 1

𝑌 2 3 1 2 8 4 9 1 4

Total 8 6 6 10 30 24 20 14 34

Step 1: 𝑁 = 16

Step 2: T=30

( )
Step 3: = = 56.25

Step 4:

TSS=∑ 𝑋 +∑𝑋 +∑𝑋 +∑𝑋 −

= 24 + 20 + 14 + 34 − 56.25

= 35.75

Step 5:

(∑ ) (∑ ) (∑ ) (∑ )
SSC= + + + − 𝑁 →Number of elements in each column

36
 (8) (6) (6) (10)
= + + + − 56.25
4 4 4 4

= 2.75

Step 6:

(∑ ) (∑ ) (∑ ) (∑ )
SSR= + + + − 𝑁 →Number of elements in each row

(8) (14) (0) (8)

= + + + − 56.25
4 4 4 4

= 24.75

Table II: To find SST:

P 0 1 2 2 5

Q 4 -1 1 2 6

R 2 3 3 1 9

S 2 3 0 5 10

(5) (6) (9) (10)

𝑆𝑆𝑇 = + + + − 56.25
4 4 4 4

= 4.25

𝑺𝑺𝑬 = 𝑻𝑺𝑺 − 𝑺𝑺𝑪 − 𝑺𝑺𝑹 − 𝑺𝑺𝑻

= 35.75 − 2.75 − 24.75 − 4.25 = 4

Step 7: ANOVA

Source of Sum of Degree of Mean squares Variance-ratio Table value at

variaions squares freedom 5% level

Between 𝑆𝑆𝐶 𝑛−1=3 𝑆𝑆𝐶 𝑀𝑆𝐶 0.917 𝐹 (3,6) = 4.76

𝑀𝑆𝐶 = 𝐹 = = = 1.375
𝑛−1 𝑀𝑆𝐸 0.667
Column = 2.75
= 0.917

37
 Between 𝑆𝑆𝑅 𝑛−1=3 𝑆𝑆𝑅 𝑀𝑆𝑅 8.25 𝐹 (3,6) = 4.76
𝑀𝑆𝑅 = 𝐹 = = = 12.369
𝑛−1 𝑀𝑆𝐸 0.667
Row = 24.75
= 8.25

Between 𝑆𝑆𝑇 𝑛−1=3 𝑆𝑆𝑇 𝑀𝑆𝑇 1.417 𝐹 (3,6) = 4.76

𝑀𝑆𝑇 = 𝐹 = = = 2.124
𝑛−1 𝑀𝑆𝐸 0.667
Treatments = 4.25
= 1.417

Error 𝑆𝑆𝐸 = 4 (𝑛 − 1) 𝑆𝑆𝐸 −

𝑀𝑆𝐸 =
6
(𝑛 − 2) = 6
= 0.667

Total 𝑇𝑆𝑆

= 35.75

Step 8: Conclusion:

Cal 𝐹 < Tab 𝐹

Cal 𝐹 > Tab 𝐹

Cal 𝐹 < Tab 𝐹

There is a significant difference between rows.

But there is no significant difference between columns and treatments.

7. The following is a Latin square design when 4 varieties of seeds are being tested. Set up

the analysis of variance table and state your conclusion. You may carry out suitable

change of obtain and scale.

A 105 B 95 C 125 D 115

C 115 D 125 A 105 B 105

D 115 C 95 B 105 A 115

B 95 A 135 D 95 C 115

38
 Solution:

Subtract 100 and then divided by 5 we get

A 1 B -1 C 5 D 3

C 3 D 5 A 1 B 1

D 3 C -1 B 1 A 3

B -1 A 7 D -1 C 3

Table I

𝒀𝒏 (𝑋 ) (𝑋 ) (𝑋 ) (𝑋 ) Total 𝑋 𝑋 𝑋 𝑋

𝑌 1 -1 5 3 8 1 1 25 9

𝑌 3 5 1 1 10 9 25 1 1

𝑌 3 -1 1 3 6 9 1 1 9

𝑌 -1 7 -1 3 8 1 49 1 9

Total 6 10 6 10 32 20 76 28 28

Step 1: 𝑁 = 16

Step 2: T=32

( )
Step 3: = = 64

Step 4:

TSS=∑ 𝑋 +∑𝑋 +∑𝑋 +∑𝑋 −

= 20 + 76 + 28 + 28 − 64

= 88

Step 5:

(∑ ) (∑ ) (∑ ) (∑ )
SSC= + + + − 𝑁 →Number of elements in each column

(6) (10) (6) (10)

= + + + − 64
4 4 4 4
39
 = 9 + 25 + 9 + 25 − 64 = 4

Step 6:

(∑ ) (∑ ) (∑ ) (∑ )
SSR= + + + − 𝑁 →Number of elements in each row

(8) (10) (6) (8)

= + + + − 64
4 4 4 4

= 16 + 25 + 9 + 16 − 64

=2

Table II: To find SST:

Total

A 1 1 3 7 12

B -1 1 1 -1 0

C 5 3 -1 3 10

D 3 5 3 -1 10

(12) (0) (10) (10) 𝑇

𝑆𝑆𝑇 = + + + −
4 4 4 4 𝑁

= 36 + 0 + 25 + 25 − 64 = 22

𝑺𝑺𝑬 = 𝑻𝑺𝑺 − 𝑺𝑺𝑪 − 𝑺𝑺𝑹 − 𝑺𝑺𝑻

= 88 − 4 − 2 − 22 = 60

Step 7: ANOVA

Source of Sum of Degree of Mean squares Variance-ratio Table value at

variaions squares freedom 5% level

Between 𝑆𝑆𝐶 = 4 𝑛−1=3 𝑆𝑆𝐶 𝑀𝑆𝐶 𝐹 (6,3) = 8.94

𝑀𝑆𝐶 = = 1.33 𝐹 =
𝑛−1 𝑀𝑆𝐸
Column
10
= = 7.52
1.33

40
 Between 𝑆𝑆𝑅 = 2 𝑛−1=3 𝑆𝑆𝑅 𝑀𝑆𝑅 𝐹 (6,3) = 8.94
𝑀𝑆𝑅 = = 0.67 𝐹 =
𝑛−1 𝑀𝑆𝐸
Row
10
= = 14.9
0.67

Between 𝑆𝑆𝑇 = 22 𝑛−1=3 𝑆𝑆𝑇 𝑀𝑆𝑇 𝐹 (6,3) = 8.94

𝑀𝑆𝑇 = = 7.33 𝐹 =
𝑛−1 𝑀𝑆𝐸
Treatments
10
= = 1.36
7.33

Error 𝑆𝑆𝐸 = 60 (𝑛 − 1)(𝑛 − 2) 𝑆𝑆𝐸 −

𝑀𝑆𝐸 =
(𝑛 − 1)(𝑛 − 2)
=6
= 10

Total 𝑇𝑆𝑆 = 88 15

Step 8: Conclusion: Cal 𝐹 > Tab 𝐹

Cal 𝐹 < Tab 𝐹

Cal 𝐹 < Tab 𝐹

There is a significant difference between rows as well as columns.

But there is no significant difference between treatments.

41
 UNIT-III

SOLUTION OF EQUATIONS AND EIGENVALUE PROBLEMS

PART – A

FIXED POINT ITERATION METHOD

1. What is the order of convergence and the condition for convergence of fixed point

iteration method?

Sol:

Order of convergence: 1

Condition for convergence:  ' ( x )  1

NEWTON’S METHOD (OR) NEWTON RAPHSON METHOD

2. State the order of convergence and condition for convergence of Newton-Raphson

method. (OR)

Write the convergence condition and order of convergence for Newton-Raphson method.

Solution: Order of convergence is two.

Condition for convergence is 𝑓(𝑥). 𝑓 ′′ (𝑥) < 𝑓 ′ (𝑥)

3. Find the smallest positive roots of the equation 𝒙𝟑 − 𝟐𝒙 + 𝟎. 𝟓 = 𝟎

Solution:

f(x) = 𝒙𝟑 − 𝟐𝒙 + 𝟎. 𝟓

42
 𝑓 ′ (𝑥) = 𝟑𝒙𝟐 − 𝟐

f(0) = 0.5(+𝑣𝑒)

f(1) = −0.5 (−𝑣𝑒)

Hence the roots lies between 0 and 1. Since the value of f(x) at x=0 is very close to zero than the

value of f(x) at x=1, we can say that the root is very close to 0. Therefore we can assume that 𝑥 = 0

is the initial approximation to the root.

Newton’s formula is

𝒇(𝒙𝒏 )
𝒙𝒏 𝟏 = 𝒙𝒏 −
𝒇′(𝒙𝒏 )

Putting n=0 in (1), we get the first approximation 𝑥 to the root, given by

( ) .
𝑥 = 𝑥 − =0−
′( )

𝑥 = 0.25

Putting n=1 in (1), we get the second approximation 𝑥 to the root, given by

( ) ( . ) ( . ) .
𝑥 = 𝑥 − = 0.25 −
′( ) ( . )

.
= 0.25 − = 0.2586
.

𝒙𝟐 = 𝟎. 𝟐𝟓𝟖𝟔

Putting n=2 in (1), we get the third approximation 𝑥 to the root, given by

( ) ( . ) ( . ) .
𝑥 = 𝑥 − = 0.2586 −
′( ) ( . )

𝒙𝟑 = 𝟎. 𝟐𝟓𝟖𝟔

Hence the smallest positive root is 𝟎. 𝟐𝟓𝟖𝟔.

43
4. Derive the formula to find the value of 𝟏 𝑵 where 𝑵 ≠ 𝟎 , using Newton Raphson method.

Solution:

Let 𝑥 =

1
𝑁=
𝑥
1
−𝑁 =0
𝑥

𝑓(𝑥) = − 𝑁 ; 𝑓 ′ (𝑥) = −

𝒇(𝒙𝒏 )
The Newton’s formula is 𝒙𝒏 𝟏 = 𝒙𝒏 −
𝒇′ (𝒙𝒏 )

−𝑁
=𝑥 −
−

1
=𝑥 + −𝑁 𝑋 𝑥
𝑥

=𝑥 +𝑥 −𝑥 𝑁

𝒙𝒏 𝟏 = 𝑥 (2 − 𝑁𝑥 )

𝟑
5. Arrive a formula to find the value of √𝑵 where 𝑵 ≠ 𝟎, using Newton-Raphson method.

Solution:

Let 𝑥 = √𝑁

𝑥 =𝑁

𝑥 −𝑁 =0

𝑓(𝑥) = 𝑥 − 𝑁 ; 𝑓 ′ (𝑥) = 3𝑥

𝐵𝑦 Newton-Raphson method

44
 𝑓(𝑥 )
𝑥 =𝑥 −
𝑓 ′ (𝑥 )

𝑥 − 𝑁 3𝑥 −𝑥 +𝑁
=𝑥 − =
3𝑥 3𝑥

1 2𝑥 + 𝑁
=
3 𝑥

1 𝑁
= 2𝑥 + , 𝑛 = 0,1,2, … ….
3 𝑥

GAUSSIAN ELIMINATION AND GAUSS – JORDON METHODS

6. Give two direct methods to solve a system of linear equation.

Solution:

* Gauss Elimination Method

* Gauss Jordon Method.

7. Compare Gauss – Jacobi and Gauss – Sedial method.

Solution:

S.No Gauss – Jacobi method Gauss – Sedial method

1. Convergence rate is slow The rate of convergence of Gauss – Seidal method is

fast, roughly twice that of Gauss – Jacobi

2. Indirect method Indirect method

3. condition for convergence is the Condition for convergence is the co-efficient matrix is

coefficient matrix is diagonally diagonally dominant.

dominant

8. Solve 3x + 2y = 4, 2x – 3y = 7 by Gauss elimination method.

45
 Solution:

Given 3x + 2y = 4

2x – 3y = 7

The given system is equivalent to

3 2 𝑥 4
=
2 −3 𝑦 7

3 2 4
Here [𝐴, 𝐵] =
2 −3 7

3 2 4
= 𝑅 ↔ 3𝑅 − 2𝑅
0 −13 13

This is an upper triangular matrix

Using backward substitution method

−13𝑦 = 13

𝑦 = −1

3𝑥 + 2𝑦 = 4

3𝑥 − 2 = 4

3𝑥 = 6

𝑥=2

Hence the solution is x = 2 and y = -1

9. Which iterative method converges faster for solving linear system of equations? Why?

Sol:

Gauss Seidal method is solving for linear system of equations converge faster. In this method the

rate of convergence is roughly twice as fast as that of Gauss- Jacobi’s method.

10.Write the uses of power method?

Sol:

To find the numerically largest eigen value of a given matrix.

EIGEN VALUE OF A MATRIX BY POWER METHOD

46
 𝟏 𝟐
11. Find the dominant eigen value and eigenvector of the matrix by power method.
𝟑 𝟒

1
Solution: 𝐿𝑒𝑡 𝑋 =
1

1 2 1 3 0.43
𝐴𝑋 = = =7 = 7𝑋
3 4 1 7 1

1 2 0.43 2.43 0.46

𝐴𝑋 = = = 5.29 = 5.29𝑋
3 4 1 5.29 1

1 2 0.46 2.46 0.46

𝐴𝑋 = = = 5.38 = 5.38𝑋
3 4 1 5.38 1

Hence the dominant eigen value=5.38

0.46
The corresponding eigen vector= .
1

PART – B

FIXED POINT ITERATION METHOD

1. Using fixed point iteration method to find the positive root of the equation

cos x  3x  1  0 .

Sol:

Given that cos 𝑥 = 3𝑥 − 1

Let 𝑓(𝑥) = 𝑐𝑜𝑠𝑥 − 3𝑥 + 1 = 0

𝑓(0) = 1 − 0 + 1 = 2 = +𝑣𝑒

𝑓(1) = 𝑐𝑜𝑠1 − 3 + 1 = −1.4597 = −𝑣𝑒

So, a root lies between 0 and 1

The given equation may be written as

1
𝑥 = (1 + 𝑐𝑜𝑠𝑥) = 𝑔(𝑥)
3
1
𝑔(𝑥) = (1 + 𝑐𝑜𝑠𝑥)
3

47
 1
𝑔′ (𝑥) = − 𝑠𝑖𝑛𝑥
3
1
|𝑔′ (𝑥)| = 𝑠𝑖𝑛𝑥
3
|𝑔′ (0)| = 0 < 1

1
|𝑔′ (1)| = sin 1 = 0.2804 < 1
3

So, the method can be applied.

Let 𝑥 = 0.6

1 1
𝑥 = [1 + 𝑐𝑜𝑠𝑥 ] = [1 + cos (0.6)] = 0.60845
3 3
1
𝑥 = [1 + 𝑐𝑜𝑠𝑥 ] = 0.60684
3
1
𝑥 = [1 + 𝑐𝑜𝑠𝑥 ] = 0.60715
3
1
𝑥 = [1 + 𝑐𝑜𝑠𝑥 ] = 0.60709
3
1
𝑥 = [1 + 𝑐𝑜𝑠𝑥 ] = 0.60710
3
1
𝑥 = [1 + 𝑐𝑜𝑠𝑥 ] = 0.60710
3

Here 𝑥 = 𝑥 = 0.60710

Hence, the better approximate root is 0.60710

2. Solve e x  3 x  0 by method of fixed point iteration.

Sol:

G.T: e x  3 x  0

Let f ( x )  e x  3 x

f (0)  e 0  3(0)  1( ve)

f (1)  e 1  3(1)  e  3(ve)

Therefore the root lies between 0 &1.

48
 ex
The given equation is of the form, x   g ( x)
3

ex
g ' ( x) 
3

ex
g ' ( x) 
3

1
g ' ( 0)  1
3

e
g ' (1)  1
3

Hence the condition is satisfied

Let us assume x 0  0.6

e x0 1 0.6
x1   e  0.6074
3 3

e x1 1 0.6074
x2   e  0.6119
3 3

e x2
x3   0.6146
3

e x3
` x4   0.6163
3

e x4
x5   0.6174
3

e x5
x6   0.6180
3

e x6
x7   0.6184
3

49
 e x7
x8   0.6187
3

e x8
x9   0.6188
3

e x9
x10   0.6189
3

e x10
x11   0.6190
3

e x11
x12   0.6190
3

 x11  x12  0.6190 correct to 4 decimal places.

Hence, the better approximate root is 0.6190.

NEWTON’S METHOD (OR) NEWTON RAPHSON METHOD

3. Solve the equation 𝒙𝒍𝒐𝒈𝟏𝟎 𝒙 = 𝟏. 𝟐 using Newton-Raphson method.

Solution:

Let 𝑓(𝑥) = 𝑥𝑙𝑜𝑔 𝑥 − 1.2 ⇒ 𝑓 ′ (𝑥) = 𝑥 × 𝑙𝑜𝑔 𝑒 + 𝑙𝑜𝑔 𝑥

𝑓 ′ (𝑥) = 𝑙𝑜𝑔 𝑒 + 𝑙𝑜𝑔 𝑥

𝑓(0) = 0 𝑙𝑜𝑔 (0) − 1.2 = −1.2 = −𝑣𝑒

𝑓(1) = 1 𝑙𝑜𝑔 (1) − 1.2 = −1.2 = −𝑣𝑒

𝑓(2) = 2 𝑙𝑜𝑔 (2) − 1.2 = −0.598 = −𝑣𝑒

𝑓(3) = 3 𝑙𝑜𝑔 (3) − 1.2 = 0.231 = +𝑣𝑒

Therefore the root lies between 2 & 3

50
 |𝑓(2)| > |𝑓(3)|

Hence the root is nearer to 3choose 𝑥 = 2.7

𝑓(𝑥 )
𝑥 =𝑥 −
𝑓′(𝑥 )

𝑓(𝑥 ) 𝑓(2.7) 2.7 𝑙𝑜𝑔 (2.7) − 1.2 −0.035

𝑥 =𝑥 − ′
= 2.7 − ′ =1− = 2.7 −
𝑓 (𝑥 ) 𝑓 (2.7) 𝑙𝑜𝑔 𝑒 + 𝑙𝑜𝑔 2.7 0.867

𝑥 = 2.740

𝑓(𝑥 ) 𝑓(2.740) −0.006

𝑥 =𝑥 − ′
= 2.740 − ′ = 2.740 −
𝑓 (𝑥 ) 𝑓 (2.740) 0.872

𝑥 = 2.741

𝑓(𝑥 ) 𝑓(2.741) −.003

𝑥 =𝑥 − ′
= 2.741 − ′ = 2.741 −
𝑓 (𝑥 ) 𝑓 (2.741) 0872

𝑥 = 2.741

We observe that the root 𝑥 = 𝑥 = 2.741Correct to 3 decimal places. Hence the required

root correct to three decimal places is 2.741

4. Find the real positive root of 𝟑𝒙 − 𝒄𝒐𝒔 𝒙 − 𝟏 = 𝟎 by Newton’s method correct to 5

decimal places.

Solution :

Let 𝑓(𝑥) = 3𝑥 − 𝑐𝑜𝑠 𝑥 − 1

𝑓 ′ (𝑥) = 3 + 𝑠𝑖𝑛 𝑥

𝑓(0) = 0 − 1 − 1 = −2 = −𝑣𝑒

𝑓(1) = 3 − 𝑐𝑜𝑠 1 − 1 = 2 − 𝑐𝑜𝑠 1 = 1.459698 = +𝑣𝑒

51
 Therefore a root lies between 0 and 1.

|𝑓(0)| > |𝑓(1)|

Hence the root lies between 0 and 1.

𝐿𝑒𝑡 𝑥 = 1

𝑓(𝑥 )
𝑥 =𝑥 − … … … (1)
𝑓 ′ (𝑥 )

Let n=0 in equation (1)

𝑓(𝑥 )
𝑥 =𝑥 −
𝑓 ′ (𝑥 )

𝑓(1)
=1−
𝑓 ′ (1)

3(1) − cos(1) − 1
=1−
3 + sin(1)

1.45970
=1−
3.84147

= 1 − 0.37998

𝑥 = 0.62002

Let n=1 in equation (1)

𝑓(𝑥 )
𝑥 =𝑥 −
𝑓 ′ (𝑥 )

𝑓(0.62002)
= 0.62002 −
𝑓 ′ (0.62002)

3(0.62002) − cos(0.62002) − 1
= 0.62002 −
3 + sin(0.62002)

𝑥 = 0.60712

Let n=2 in equation (1)

𝑓(𝑥 )
𝑥 =𝑥 −
𝑓 ′ (𝑥 )

𝑓(0.60712)
= 0.60712 −
𝑓 ′ (0.607102)

52
 3(0.60712) − cos(0.60712) − 1
= 0.60712 −
3 + sin(0.60712)

𝑥 = 0.60710

Let n=3 in equation (1)

𝑓(𝑥 )
𝑥 =𝑥 −
𝑓 ′ (𝑥 )

𝑓(0.60710)
= 0.60710 −
𝑓 ′ (0.60710)

3(0.60710) − cos(0.60710) − 1
= 0.60710 −
3 + sin(0.60710)

𝑥 = 0.60710

From 𝑥 and 𝑥 we findout the root is 0.60710 correct to five decimal places.

5. Interpret the Newton’s iterative formula to calculate the reciprocal of N and hence find

the value of 1/26.

Sol:

Let 𝑥 =

1
𝑁=
𝑥
1
−𝑁 =0
𝑥

𝑓(𝑥) = − 𝑁 ; 𝑓 ′ (𝑥) = −

𝒇(𝒙𝒏 )
The Newton’s formula is 𝒙𝒏 𝟏 = 𝒙𝒏 −
𝒇′ (𝒙𝒏 )

−𝑁
=𝑥 −
−

1
=𝑥 + −𝑁 𝑋 𝑥
𝑥

53
 =𝑥 +𝑥 −𝑥 𝑁

= 𝑥 (2 − 𝑁𝑥 )

To find 1/26, take N=26

Let x 0  0.04

W.K.T 𝒙𝒏 𝟏 = 𝑥 (2 − 𝑁𝑥 )

𝒙𝟏 = 𝑥 (2 − 26𝑥 )

= 0.04(2 − 26(0.04))

=0.0384

𝒙𝟐 = 𝑥 (2 − 26𝑥 )

= 0.0384(2 − 26(0.0384))

=0.0385

𝒙𝟑 = 𝑥 (2 − 26𝑥 )

= 0.0385(2 − 26(0.0385))

=0.0385

Here 𝒙𝟐 = 𝑥 = 0.0385

Hence the value of 1/26=0.0385

SOLUTION OF LINEAR SYSTEM BY GAUSSIAN ELIMINATION METHOD

6. Solve the system of equations using Gauss elimination method

𝟓𝒙 − 𝟐𝒚 + 𝒛 = 𝟒; 𝟕𝒙 + 𝒚 − 𝟓𝒛 = 𝟖; 𝟑𝒙 + 𝟕𝒚 + 𝟒𝒛 = 𝟏𝟎 .
Solution:
The given system is equivalent to

54
 5 −2 1 𝑥 4
7 1 −5 𝑦 = 8
3 7 4 𝑧 10
5 −2 1 10
[𝐴, 𝐵] = 7 1 −5 18
3 7 4 16

5 −2 1 4 ⬚ 𝑅 ↔ 7𝑅 − 5𝑅 ⬚
~ 0 −19 27 −12
⬚ 𝑅 ↔ 3𝑅 − 5𝑅 ⬚
0 −41 −17 −38

5 −2 1 4
⬚ ⬚ ⬚
~ 0 −19 27 −12
⬚ 𝑅 ↔ 41𝑅 − 19𝑅 ⬚
0 0 1430 230

Use back substitution to find the solution to the system.

1430𝑧 = 230

𝑧 = 230/1430

𝑧 = 0.161

−19𝑦 + 27𝑧 = −12

−19𝑦 = −12 − 4.343

𝑌 = 0.860

5𝑥 − 2𝑦 + 𝑧 = 4

5𝑥 − 1.72 + 1.161 = 4

𝑥 = 1.112

Hence 𝒙 = 𝟏. 𝟏𝟏𝟐, 𝒚 = 𝟎. 𝟖𝟔𝟎, 𝒛 = 𝟎. 𝟏𝟔𝟏.

7. Solve the following equations by Gauss elimination method:

𝟐𝒙 + 𝒚 + 𝟒𝒛 = 𝟏𝟐; 𝟖𝒙 − 𝟑𝒚 + 𝟐𝒛 = 𝟐𝟎; 𝟒𝒙 + 𝟏𝟏𝒚 − 𝒛 = 𝟑𝟑

Solution:
The given system is equivalent to

2 1 4 𝑥 12
8 −3 2 𝑦 = 20
1 4 9 𝑧 33
55
 2 1 4 12
[𝐴, 𝐵] = 8 −3 2 20
1 4 9 33

2 1 4 12
~ 0 7 14 18 𝑅 → 4𝑅 − 𝑅 , 𝑅 → 2𝑅 − 𝑅
0 −9 9 −9

2 1 4 12
~ 0 7 14 18 𝑅 → 7𝑅 + 9𝑅
0 0 189 189

Use back substitution to find the solution to the system.

189𝑧 = 189

𝒛=𝟏

7𝑦 + 14𝑧 = 28 ⇒ 7𝑦 + 14 = 28 ⇒ 7 𝑦 = 14

𝒚=𝟐

2𝑥 + 𝑦 + 4𝑧 = 12 ⇒ 2𝑥 + 2 + 4 = 12

2𝑥 = 6

𝒙=𝟑

Hence 𝒙 = 𝟑, 𝒚 = 𝟐, 𝒛 = 𝟏.

SOLUTION OF LINEAR SYSTEM BY GAUSS – JORDAN METHODS

8. Using the Gauss – Jordan method solve the following equations 10x + y + z = 12,

2x + 10y + z = 13, x + y +5z =7

Solution:

Given 10𝑥 + 𝑦 + 𝑧 = 12

2𝑥 + 10𝑦 + 𝑧 = 13

𝑥 + 𝑦 + 5𝑧 = 7

Interchanging the first and the last equation then

56
 𝑥 + 𝑦 + 5𝑧 = 7

2𝑥 + 10𝑦 + 𝑧 = 13

10𝑥 + 𝑦 + 𝑧 = 12

The given system is equivalent to

1 1 5 𝑥 7
2 10 1 𝑦 = 13
10 1 1 𝑧 12

AX = B

1 1 5 7
Here [𝐴, 𝐵] = 2 10 1 13
10 1 1 12

Fix the pivot element row and make the other elements zero in the pivot element column.

1 1 5 7
~ 0 8 −9 −1 𝑅 ↔ 𝑅 − 2𝑅 & 𝑅 ↔ 𝑅 − 10𝑅
0 −9 −49 −58

8 0 49 57
~ 0 8 −9 −1 𝑅 ↔ 8𝑅 − 𝑅 & 𝑅 ↔ 8𝑅 + 9𝑅
0 0 473 473

8 0 49 57
~ 0 8 −9 −1 𝑅 ↔
0 0 1 1
8 0 08
~ 0 8 0 8 𝑅 ↔ 𝑅 − 49𝑅 & 𝑅 ↔ 𝑅 + 9𝑅
0 0 11
1 0 01
~ 0 1 01 𝑅 ↔ &𝑅 ↔
0 0 11

Therefore the solution is 𝒙 = 𝟏, 𝒚 = 𝟏, 𝒛 = 𝟏

9. Using the Gauss – Jordan method solve the following equations 𝟐𝒙 − 𝒚 + 𝟑𝒛 = 𝟖,

−𝒙 + 𝟐𝒚 + 𝒛 = 𝟒, 𝟑𝒙 + 𝒚 − 𝟒𝒛 = 𝟎

Solution:

Given 2𝑥 − 𝑦 + 3𝑧 = 8

−𝑥 + 2𝑦 + 𝑧 = 4

3𝑥 + 𝑦 − 4𝑧 = 0

The given system is equivalent to

57
 2 −1 3 𝑥 8
−1 2 1 𝑦 = 4
3 1 −4 𝑧 0

𝐴𝑋 = 𝐵

2 −1 3 8
Here [𝐴, 𝐵] = −1 2 1 4
3 1 −4 0

Fix the pivot element row and make the other elements zero in the pivot element column.

2 −1 3 8
~ 0 3 5 16 𝑅 ↔ 2𝑅 + 𝑅 & 𝑅 ↔ 2𝑅 − 3𝑅
0 5 −17 −24
6 0 14 40 𝑅 ↔ 3𝑅 + 𝑅
~ 0 3 5 16
𝑅 ↔ 3𝑅 − 5𝑅
0 0 −76 −152
6 0 14 40
~ 0 3 5 16 𝑅 ↔
0 0 1 2

6 0 0 12
~ 0 3 0 6 𝑅 ↔ 𝑅 − 14𝑅 & 𝑅 ↔ 𝑅 − 5𝑅
0 0 1 2
1 0 02 𝑅 𝑅
~ 0 1 02 𝑅 ↔ &𝑅 ↔
6 3
0 0 12

Therefore the solution is 𝒙 = 𝟐, 𝒚 = 𝟐, 𝒛 = 𝟐

GAUSS – JACOBI METHOD AND GAUSS – SEDIAL METHOD

10. Solve the system of equation by Gauss – Sedial method correct to 4 decimal places

20x + y –2 z = 17, 3x + 20y -z = -18, 2x -3y +20z = 25

Solution: .

Given 20x + y –2 z = 17

3x + 20y -z = -18

2x -3y +20z = 25

As the coefficient matrix is diagonally dominant solving for x, y, z we get

58
 𝑥= [17 − 𝑦 + 2𝑧], 𝑦 = [−18 − 3𝑥 + 𝑧], 𝑧 = [25 − 2𝑥 + 3𝑦]

Let the initial value be y=0, z=0

Iteration 17 − 𝑦 + 2𝑧 −18 − 3𝑥 + 𝑧 25 − 2𝑥 + 3𝑦
𝑥= 𝑦= 𝑧=
20 20 20

1 0.85 -1.0275 1.0109

2 1.0025 -0.9998 0.9998

3 1.0000 -1.0000 1.0000

4 1 -1 1

Hence x = 1, y = -1, z = 1.

11. Solve the system of equation by Gauss – Seidel method 28x +4 y – z = 32, x + 3y +10z = 24,

2x +17y +4z = 35.

Solution: .

Given 28𝑥 + 4 𝑦 – 𝑧 = 32

𝑥 + 3𝑦 + 10𝑧 = 24

2𝑥 + 17𝑦 + 4𝑧 = 35

As the coefficient matrix is diagonally dominant solving for x, y, z we get

𝑥= [32 − 4𝑦 + 𝑧]

𝑦= [35 − 2𝑥 − 4𝑧]

𝑧= [24 − 𝑥 − 3𝑦]

Let the initial value be y=0, z=0

Iteration 32 − 4𝑦 + 𝑧 35 − 2𝑥 − 4𝑧 24 − 𝑥 − 3𝑦
𝑥= 𝑦= 𝑧=
28 17 20

59
 Let1the initial value be y=0, z=0
1.1429 1.9244 1.8084

2 0.9325 1.5236 1.8497

3 0.9913 1.5070 1.8488

4 0.9936 1.5069 1.8486

5 0.9936 1.5069 1.8486

Hence x =0.9936 , y = 1.5069, z = 1.8486

EIGEN VALUE OF A MATRIX BY POWER METHOD

𝟏 𝟔 𝟏
12. Find the largest Eigen value and the corresponding Eigen vector of 𝑨 = 𝟏 𝟐 𝟎 using
𝟎 𝟎 𝟑

power method. Using 𝒙𝟏 = (𝟏 𝟎 𝟎)𝑻 as initial vector.

Solution:

1
Let 𝑋 = 0 be an approximate eigen value.
0

1 6 1 1 1 1
𝐴𝑋 = 1 2 0 0 = 1 =1 1 =1𝑋
0 0 3 0 0 0

1 6 1 1 7 1
𝐴𝑋 = 1 2 0 1 = 3 = 7 0.4286 = 7 𝑋
0 0 3 0 0 0

1 6 1 1 3.5714 1
𝐴𝑋 = 1 2 0 0.4286 = 1.8572 = 3.5714 0.52 = 3.5714 𝑋
0 0 3 0 0 0

1 6 1 1 4.12 1
𝐴𝑋 = 1 2 0 0.52 = 2.04 = 4.12 0.4951 = 4.12 𝑋
0 0 3 0 0 0
1 6 1 1 3.9706 1
𝐴𝑋 = 1 2 0 0.4951 = 1.9902 = 3.9706 0.5012 = 3.9706 𝑋
0 0 3 0 0 0

60
 1 6 1 1 4.0072 1
𝐴𝑋 = 1 2 0 0.5012 = 2.0024 = 4.0072 0.4997 = 4.0072 𝑋
0 0 3 0 0 0
1 6 1 1 3.9982 1
𝐴𝑋 = 1 2 0 0.4997 = 1.9994 = 3.9982 0.5000 = 3.9982 𝑋
0 0 3 0 0 0

1 6 1 1 4 1
𝐴𝑋 = 1 2 0 0.5 = 2 = 4 0.5 = 4 𝑋
0 0 3 0 0 0
1 6 1 1 4 1
𝐴𝑋 = 1 2 0 0.5 = 2 = 4 0.5
0 0 3 0 0 0

Therefore Dominant eigen value =4; corresponding eigen vector is (1, 0.5, 0)

13. Find , by power method, the largest Eigen value and the corresponding Eigen vector of a

1 3 −1
matrix 𝑨 = 3 2 4 with initial vector ( 𝟏 𝟏 𝟏)𝑻 .
−1 4 10

Solution:

1
Let 𝑋 = 1 be an arbitrary initial eigen vector.
1

1 3 −1 1 3 0.231
𝐴𝑋 = 3 2 4 1 = 9 = 13 0.692 = 13𝑋
−1 4 10 1 13 1

1 3 −1 0.231 1.307 0.104

𝐴𝑋 = 3 2 4 0.692 = 6.077 = 12.537 0.485 = 12.537𝑋
−1 4 10 1 12.537 1

1 3 −1 0.104 0.559 0.047

𝐴𝑋 = 3 2 4 0.485 = 5.282 = 11.836 0.485 = 11.836𝑋
−1 4 10 1 11.836 1

1 3 −1 0.047 0.385 0.033

𝐴𝑋 = 3 2 4 0.485 = 5.033 = 11.737 0.429 = 11.737𝑋
−1 4 10 1 11.737 1
1 3 −1 0.033 0.32 0.027
𝐴𝑋 = 3 2 4 0.429 = 4.957 = 11.683 0.424 = 11.683𝑋
−1 4 10 1 1 1

1 3 −1 0.027 0.299 0.026

𝐴𝑋 = 3 2 4 0.424 = 4.929 = 11.669 0.422 = 11.669𝑋
−1 4 10 1 11.669 1
61
 1 3 −1 0.026 0.299 0.025
𝐴𝑋 = 3 2 4 0.422 = 4.922 = 11.662 0.422 = 11.662𝑋
−1 4 10 1 11.662 1

1 3 −1 0.025 0.291 0.025

𝐴𝑋 = 3 2 4 0.422 = 4.919 = 11.663 0.422
−1 4 10 1 11.663 1

0.025
Therefore, the dominant eigenvector is 0.422 , eigenvalue is 11.663.
1

UNIT-IV

INTERPOLATION, NUMERICAL DIFFERENTIATION AND NUMERICAL

INTEGRATION

PART-A

LAGRANGE’S INTERPOLATION

1. Write down the Lagrange’s Interpolation formula.

Solution:

Let y = f(x) be a function which takes the values 𝑦 , 𝑦 , 𝑦 , … . . 𝑦 corresponding to

𝑥 ,𝑥 ,𝑥 ,….𝑥

Then Lagrange’s interpolation formula is

( )( )…( ) ( )( )…( )
𝑦 = 𝑓(𝑥) = ( )( )…(
𝑦 +( )( )…(
𝑦
) )

(𝑥 − 𝑥 )(𝑥 − 𝑥 ) … (𝑥 − 𝑥 )
+ ……+ 𝑦
(𝑥 − 𝑥 )(𝑥 − 𝑥 ) … (𝑥 − 𝑥 )

2. Find the second degree polynomial through the points (0,2),(2,1),(1,0) using Lagrange’s

formula.

Solution:

We use Lagrange’s interpolation formula

62
 (𝑥 − 𝑥 )(𝑥 − 𝑥 ) (𝑥 − 𝑥 )(𝑥 − 𝑥 )
𝑦 = 𝑓(𝑥) = 𝑦 + 𝑦
(𝑥 − 𝑥 )(𝑥 − 𝑥 ) (𝑥 − 𝑥 )(𝑥 − 𝑥 )

(𝑥 − 𝑥 )(𝑥 − 𝑥 )
+ 𝑦
(𝑥 − 𝑥 )(𝑥 − 𝑥 )

(𝑥 − 2)(𝑥 − 1) (𝑥 − 0)(𝑥 − 1) (𝑥 − 0)(𝑥 − 2)

= .2 + .1 + .0
(0 − 2)(0 − 1) (2 − 0)(2 − 1) (1 − 0)(1 − 2)

1 1
= 𝑥 − 3𝑥 + 2 + (𝑥 − 𝑥) = (2𝑥 − 6𝑥 + 4 + 𝑥 − 𝑥)
2 2
1
𝑦 = (3𝑥 − 7𝑥 + 4)
2

DIVIDED DIFFERENCES

3. Distinguish between interpolation and extrapolation.

Solution:

Interpolation Extrapolation

To find the values of a function inside a To find the values of a function outside a

given range is interpolation. given range is extrapolation.

4. Find the divided difference of f(x) which takes the values 1, 4, 40, 85 with arguments 0,

1, 3, 4

Solution:

The divided difference table is as follows

x f(x) ∆𝑓(𝑥) ∆ 𝑓(𝑥) ∆ 𝑓(𝑥)

63
 0 1

4−1
=3
1−0
18 − 3
=5
3−0
1 4
40 − 4 = 18
.
3−1 = 0.44

45 − 18
= 6.75
3 40 4−0

85 − 40
= 45
4−3

4 85

5. Find the divided differences of 𝒇(𝒙) = 𝒙𝟑 + 𝒙 + 𝟐 for the arguments 1, 3 , 6, 11.

Solution:

𝑓(1) = 1 + 1 + 2 = 4

𝑓(3) = 3 + 3 + 2 = 32

𝑓(6) = 6 + 6 + 2 = 224

𝑓(11) = 11 + 11 + 2 = 1344

The divided difference table is as follows

x f(x) ∆𝑓(𝑥) ∆ 𝑓(𝑥) ∆ 𝑓(𝑥)

64
 1 4

32 − 4
= 14
3−1
64 − 14
= 10
6−1
3 32
224 − 32 = 64
6−3 =1

224 − 64
= 20
6 224 11 − 3

1344 − 224
= 224
11 − 6

11 1344

NEWTON’S FORWARD AND BACKWARDINTERPOLATION

6. Derive Newton’s backward interpolation formula using operator method.

(OR) State Newton’s backward formula for interpolation.

State Newton’s backward difference formula.

Solution:

𝑣(𝑣 + 1) 𝑣(𝑣 + 1)(𝑣 + 2) 𝑣(𝑣 + 1)(𝑣 + 2)(𝑣 + 3)

𝑦(𝑥) = 𝑦 + 𝑣∇𝑦 + ∇ 𝑦 + ∇ 𝑦 + ∇ 𝑦 +⋯
2! 3! 4!

Where 𝑣 =

7. Derive Newton’s forward interpolation formula using equal intervals .

Solution:

65
 𝑛(𝑛 − 1) 𝑛(𝑛 − 1)(𝑛 − 2)
𝑦 = 𝑓(𝑥 + 𝑛h) = 𝑦 + 𝑛∆𝑦 + ∆ 𝑦 + ∆ 𝑦 + +⋯
2! 3!

8. Find the first and second divided difference with arguments 𝒂, 𝒃, 𝒄 of the function

𝟏
𝒇(𝒙) =
𝒙

Solution:

If 𝑓(𝑥) = ⇒ 𝑓(𝑎) =

( ) ( )
𝑓(𝑎, 𝑏) = ∆ = =− ∵ 𝑓(𝑥 , 𝑥 ) =

𝑓(𝑏, 𝑐) − 𝑓(𝑎, 𝑏) − − 1 𝑐−𝑎 1

𝑓(𝑎, 𝑏, 𝑐) = = = =
𝑐−𝑎 𝑐−𝑎 𝑎𝑏𝑐 𝑐 − 𝑎 𝑎𝑏𝑐

∴ ∆ =

DIFFERENTION USING INTERPOLATION FORMULA

9. Write down the expression for 𝒂𝒏𝒅 𝒂𝒕 𝒙 = 𝒙𝒏 by Newton’s backward difference

formula.

Solution:

𝑑𝑦 1 1 1 1
= ∇𝑦 + ∇ 𝑦 + ∇ 𝑦 + ∇ 𝑦 + ⋯
𝑑𝑥 ℎ 2 3 4

𝑑 𝑦 1 11
= ∇ 𝑦 −∇ 𝑦 + ∇ 𝑦 +⋯
𝑑𝑥 ℎ 12

NUMERICAL INTEGRATION BY TRAPEZOIDAL METHOD

b
10. State Trapezoidal rule to evaluate  f ( x)dx .
a

Solution:

66
 b
h
 f ( x)dx  2  y
a
0  y n   2 y1  y 2  ...  y n 1 

𝟐 𝒅𝒙
11. Taking 𝒉 = 𝟎. 𝟓,evaluate ∫𝟏 using Trapezoidal rule.
𝟏 𝒙𝟐

Solution:

Here 𝑦(𝑥) =

Length of the interval = 1

x : 1 1.5 2

𝑦= : 0.5 0.3077 0.2

𝒙𝟐

ℎ = 0.5

By Trapezoidal rule

Trapezoidal rule

ℎ
= [𝑠𝑢𝑚 𝑜𝑓 𝑡ℎ𝑒 𝑓𝑖𝑟𝑠𝑡 𝑎𝑛𝑑 𝑙𝑎𝑠𝑡 𝑜𝑟𝑑𝑖𝑛𝑎𝑡𝑒𝑠]
2

+ 2[𝑠𝑢𝑚 𝑜𝑓 𝑡ℎ𝑒 𝑟𝑒𝑚𝑎𝑖𝑛𝑖𝑛𝑔 𝑜𝑟𝑑𝑖𝑛𝑎𝑡𝑒𝑠]

𝑑𝑥 ℎ
= [(0.5 + 0.2) + 2(0.3077)]
1+𝑥 2

𝑑𝑥 0.5
= [0.7 + 0.6154]
1+𝑥 2

𝑑𝑥 0.5
= [1.3154] = 0.3289
1+𝑥 2

67
 
12. Using Trapezoidal rule, evaluate  sin x dx by dividing the range into 6 equal parts.
0

Solution:


Given:  sin x dx
0

Range = b - a =   0  


Here h 
6

x 0  2 3 4 5 
6 6 6 6 6

y  sin x 0.0 0.5 0.866 1 0.866 0.5 0

(i) By Trapezoidal Rule:

b
h
 f ( x)dx  2  A  2 B 
a


= 0  0   20.5  0.866  1  0.8666  0.5
12

= 0.6220 

1
1
13. Evaluate  x dx
1
by Trapezoidal rule, dividing the range into 4equal parts.

Solution:

Here, ℎ= = ; 𝑦=

𝑥: 1/2 = 4/8 5/8 6/8 7/8 8/8

𝑓(𝑥): 1/𝑥 8/4 8/5 8/6 8/7 8/8

68
A= Sum of the first and last ordinates = + = 3

B= Sum of the remaining ordinates =8/5+8/6+8/7=856/210

1
1 h 1 856  2 
  dx  A  2 B    3    0.6971
1 x 2 16  210 
2

𝟏 𝒅𝒙
14. Evaluate ∫𝟎 using Trapezoidal rule.
𝟏 𝒙𝟐

Solution:

Here 𝑦(𝑥) =

Length of the interval = 1

𝑥 : 0 0.2 0.4 0.6 0.8 1

𝑦= : 1 0.96154 0.86207 0.73529 0.60976 0.5

𝒙𝟐

ℎ = 0.2

By Trapezoidal rule

Trapezoidal rule

ℎ
= [𝑠𝑢𝑚 𝑜𝑓 𝑡ℎ𝑒 𝑓𝑖𝑟𝑠𝑡 𝑎𝑛𝑑 𝑙𝑎𝑠𝑡 𝑜𝑟𝑑𝑖𝑛𝑎𝑡𝑒𝑠]
2

+ 2[𝑠𝑢𝑚 𝑜𝑓 𝑡ℎ𝑒 𝑟𝑒𝑚𝑎𝑖𝑛𝑖𝑛𝑔 𝑜𝑟𝑑𝑖𝑛𝑎𝑡𝑒𝑠]

∫ = [(𝑦 + 𝑦 ) + 2(𝑦 + 𝑦 + 𝑦 + 𝑦 +𝑦 )]

𝑑𝑥 0.2
= [(1 + 0.5) + 2(0.96154 + 0.86207 + 0.9412 + 0.73529 + 0.60976)]
1+𝑥 2

𝑑𝑥 0.2
= [7.83732] = 0.783732 … … . . (1)
1+𝑥 2

By actual integration,

𝑑𝑥 𝜋
= (𝑡𝑎𝑛 𝑥) = 𝑡𝑎𝑛 1 − 𝑡𝑎𝑛 0= … … (2)
1+𝑥 4

69
 𝜋
𝐹𝑟𝑜𝑚 (1)& (2) = 0.783732
4

𝜋 = 3.13493(𝑎𝑝𝑝𝑟𝑜𝑥𝑖𝑚𝑎𝑡𝑒𝑙𝑦

NUMERICAL INTEGRATION BY SIMPSON’S 1/3 AND 3/8 RULES

15. State Simpson’s one-third rule.

Solution:

Simpson’s one third rule is

ℎ
𝑓(𝑥) 𝑑𝑥 = [(𝑦 + 𝑦 ) + 4(𝑦 + 𝑦 + ⋯ + 𝑦 ) + 2(𝑦 + 𝑦 + ⋯ + 𝑦 )]
3

b d
16. State Trapezoidal rule for evaluating   f ( x, y )
a c
dx dy .

Solution:

hk
I [( Sum of values of f at Four corners) + 2(Sum of the values of f at remaining nodes
4
on the boundary )  4  sum of values of f at interior nodes 

PART-B

LAGRANGE’S INTERPOLATION

1. Find the interpolation polynomial f(x) by Lagrange’s formula and hence find f(3) for

(0,2),(1,3),(2,12)and (5,147). (OR)

Find the polynomial 𝒇(𝒙) by using Lagrange’s formula and hence find 𝒇(𝟑) for
70
 𝒙 0 1 2 5

𝒇(𝒙) 2 3 12 147

Solution:

By Lagrange’s interpolation formula, we have

(𝑥 − 𝑥 )(𝑥 − 𝑥 )(𝑥 − 𝑥 ) (𝑥 − 𝑥 )(𝑥 − 𝑥 )(𝑥 − 𝑥 )

𝑦 = 𝑓(𝑥) = 𝑦 + 𝑦
(𝑥 − 𝑥 )(𝑥 − 𝑥 )(𝑥 − 𝑥 ) (𝑥 − 𝑥 )(𝑥 − 𝑥 )(𝑥 − 𝑥 )

(𝑥 − 𝑥 )(𝑥 − 𝑥 )(𝑥 − 𝑥 ) (𝑥 − 𝑥 )(𝑥 − 𝑥 )(𝑥 − 𝑥 )

+ 𝑦 + 𝑦
(𝑥 − 𝑥 )(𝑥 − 𝑥 )(𝑥 − 𝑥 ) (𝑥 − 𝑥 )(𝑥 − 𝑥 )(𝑥 − 𝑥 )

(𝑥 − 1)(𝑥 − 2)(𝑥 − 5) (𝑥 − 0)(𝑥 − 2)(𝑥 − 5)

𝑦 = 𝑓(𝑥) = (2) + (3)
(0 − 1)(0 − 2)(0 − 5) (1 − 0)(1 − 2)(1 − 5)

(𝑥 − 0)(𝑥 − 1)(𝑥 − 5) (𝑥 − 0)(𝑥 − 1)(𝑥 − 2)

+ (12) + (147)
(2 − 0)(2 − 1)(2 − 5) (5 − 0)(5 − 1)(5 − 2)

(𝑥 − 1)(𝑥 − 2)(𝑥 − 5) 𝑥(𝑥 − 2)(𝑥 − 5)

= (2) + (3)
(−10) 4

(𝑥 − 1)(𝑥 − 5) 𝑥(𝑥 − 1)(𝑥 − 2)

+ (12) + (147)
−6 60

Put x = 3 we get

(3 − 1)(3 − 2)(3 − 5) 3(3 − 2)(3 − 5)

𝑦 = 𝑓(3) = (2) + (3)
−10 4

3(3 − 1)(3 − 5) 3(3 − 1)(3 − 2)

+ (12) + (147)
−6 60

2(−2) 3(−2) 3(2)(−2) 3(2)

= (2) + (3) + (12) + (147)
(−10) 4 (−6) 60

4 6 1 8 18 147
𝑦 = 𝑓(3) = (2) − (3) + 2(12) + (147) = − + 24 +
10 4 10 10 4 10

𝑓(3) = 35

2. Use Lagrange’s formula to construct a polynomial which takes the values

f (0)  12, f (1)  0, f (3)  6 and f ( 4)  12. Hence find f ( 2).

Solution:
71
By Lagrange’s interpolation formula, we have

(𝑥 − 𝑥 )(𝑥 − 𝑥 )(𝑥 − 𝑥 ) (𝑥 − 𝑥 )(𝑥 − 𝑥 )(𝑥 − 𝑥 )

𝑦 = 𝑓(𝑥) = 𝑦 + 𝑦
(𝑥 − 𝑥 )(𝑥 − 𝑥 )(𝑥 − 𝑥 ) (𝑥 − 𝑥 )(𝑥 − 𝑥 )(𝑥 − 𝑥 )

(𝑥 − 𝑥 )(𝑥 − 𝑥 )(𝑥 − 𝑥 ) (𝑥 − 𝑥 )(𝑥 − 𝑥 )(𝑥 − 𝑥 )

+ 𝑦 + 𝑦
(𝑥 − 𝑥 )(𝑥 − 𝑥 )(𝑥 − 𝑥 ) (𝑥 − 𝑥 )(𝑥 − 𝑥 )(𝑥 − 𝑥 )

(𝑥 − 1)(𝑥 − 3)(𝑥 − 4)
𝑦 = 𝑓(𝑥) = (−12) + 0
(0 − 1)(0 − 3)(0 − 4)

(𝑥 − 0)(𝑥 − 1)(𝑥 − 4) (𝑥 − 0)(𝑥 − 1)(𝑥 − 3)

+ (6) + (12)
(3 − 0)(3 − 1)(3 − 4) (4 − 0)(4 − 1)(4 − 3)

(𝑥 − 1)(𝑥 − 3)(𝑥 − 4) 𝑥(𝑥 − 1)(𝑥 − 4)

= (−12) + (6)
(−12) (−6)

𝑥(𝑥 − 1)(𝑥 − 3)
+ (12)
12

= (𝑥 − 1)(𝑥 − 3)(𝑥 − 4) − 𝑥(𝑥 − 1)(𝑥 − 4) + 𝑥(𝑥 − 1)(𝑥 − 3)

= (𝑥 − 1)[𝑥 − 3𝑥 − 4𝑥 + 12 − 𝑥 + 4𝑥 + 𝑥 − 3𝑥]

= (𝑥 − 1)(𝑥 − 6𝑥 + 12)

= 𝑥 − 6𝑥 + 12𝑥 − 𝑥 + 6𝑥 − 12

𝑓(𝑥) = 𝑥 − 7𝑥 + 18𝑥 − 12

∴ 𝑓(2) = 2 − 7(2) + 18(2) − 12 = 4

∴ 𝑓(2) = 4

72
 DIVIDED DIFFERENCES

3. Determine 𝒇(𝒙) as a polynomial in 𝒙 for the following data, using Newton’s divided

difference formula. Also find 𝒇(𝟑)

𝒙 -4 -1 0 2 5

𝒇(𝒙) 1245 33 5 9 1335

Solution:

𝑥 𝑓(𝑥) ∆𝑓(𝑥) ∆ 𝑓(𝑥) ∆ 𝑓(𝑥) ∆ 𝑓(𝑥)

-4 1245

33 − 124
= −404
−1 − (−4)

-1 33 −28 − (−404)
= 94
0 − (−4)

5 − 33
= −28 10 − 94
0 − (−1)
= −14
2 − (−4)

2 − (−28)
= 10
0 5 2 − (−1)

9−5 13 + 14
=2 =3
2−0 5 − (−4)

88 − 10
= 13
442 − 2 5 − (−1)
= 88
5−0
2 9
1335 − 9
= 442
5−2

5 1335

73
 By Newton’s divided difference interpolation formula,

𝑓(𝑥) = 𝑓(𝑥 ) + (𝑥 − 𝑥 )𝑓(𝑥 , 𝑥 ) + (𝑥 − 𝑥 )(𝑥 − 𝑥 )𝑓(𝑥 , 𝑥 , 𝑥 ) + (𝑥 − 𝑥 )(𝑥 − 𝑥 )(𝑥 − 𝑥 )𝑓(𝑥 , 𝑥 , 𝑥 , 𝑥 )

↳ ⋯ … … … . (1)

Here 𝑓(𝑥 ) = 1245, 𝑓(𝑥 , 𝑥 ) = −404, 𝑓(𝑥 , 𝑥 , 𝑥 ) = 94, 𝑓(𝑥 , 𝑥 , 𝑥 , 𝑥 ) = −14 & 𝑓(𝑥 , 𝑥 , 𝑥 , 𝑥 , 𝑥 ) =3,

Hence we using this formula in equation (1) we get

𝑓(𝑥) = 1245 + (𝑥 + 4)(−404) + (𝑥 + 4)(𝑥 + 1)(94) + (𝑥 + 4)(𝑥 + 1)(𝑥 − 0)(−14)

+ (𝑥 + 4)(𝑥 + 1)(𝑥 − 0)(𝑥 − 2)(3)

= 1245 − 404𝑥 − 1616 + 94𝑥 + 470𝑥 + 376 − 14𝑥 − 70𝑥 − 56𝑥

+ 3𝑥[𝑥 − 2𝑥 + 5𝑥 − 10𝑥 + 4𝑥 − 8 ]

= −14𝑥 + 24𝑥 + 10𝑥 + 5 + 3𝑥[𝑥 ∓ 3𝑥 − 6𝑥 − 8 ]

= −14𝑥 + 24𝑥 + 10𝑥 + 5 + 3𝑥 + 9𝑥 − 18𝑥 − 24𝑥

∴ 𝑓(𝑥) = 3𝑥 − 5𝑥 + 6𝑥 − 14𝑥 + 5

⇒ 𝑓(3) = 3 × 3 − 5 × 3 + 6 × 3 − 14 × 3 + 5 = 125

∴ 𝑓(3) = 125

4. Use Newton’s divided difference formula find f(9) given the values (5,150), (7,392),

(13,2366) and (17,5202)

Solution:

x f(x) ∆𝑓(𝑥) ∆ 𝑓(𝑥) ∆ 𝑓(𝑥)

74
 5 150

392 − 150
= 121
7−5
329 − 121
= 26
13 − 5
7 392
2366 − 392 = 329
13 − 7 38 − 26
=1
17 − 5
709 − 329
= 38
13 2366 17 − 7

5202 − 2366
= 709
17 − 13

17 5202

By Newton’s divided difference formula

𝑓(𝑥) = 𝑓(𝑥 ) + (𝑥 − 𝑥 )𝑓(𝑥 , 𝑥 ) + (𝑥 − 𝑥 )(𝑥 − 𝑥 )𝑓(𝑥 , 𝑥 , 𝑥 )

+ (𝑥 − 𝑥 )(𝑥 − 𝑥 )(𝑥 − 𝑥 )𝑓(𝑥 , 𝑥 , 𝑥 , 𝑥 ) … … (1)

= 150 + (𝑥 − 5)(121) + (𝑥 − 5)(𝑥 − 7)(26) + (𝑥 − 5)(𝑥 − 7)(𝑥 − 13)(1)

𝑓(9) = 150 + (9 − 5)(121) + (9 − 5)(9 − 7)(26) + (9 − 5)(9 − 7)(9 − 13)(1)

𝑓(9) = 150 + 484 + 192 − 32

𝑓(9) = 794

NEWTON’S FORWARD AND BACKWARD INTERPOLATION

5. Find a polynomial of degree two for the data by Newton’s forward difference formula.

X 0 1 2 3 4 5 6 7

y 1 2 4 7 11 16 22 29

Solution:

75
 x y ∆𝑦 ∆ 𝑦 ∆ 𝑦

0 1

1 2 1

2 0

2 4 1

3 0

3 7 1

4 0

4 11 1

5 0

5 16 1

6 0

6 22 1

7 29

Here 𝑥 = 0, 𝑦 = 1, ℎ = 1

( )
𝑦(𝑥) = 𝑦 + 𝑢∆𝑦 + ∆ 𝑦 +……
!

Where 𝑢 = = =𝑥⇒𝑢=𝑥

( )
𝑦(𝑥) = 1 + 𝑥(1) + (1)
!

=1+𝑥+ =

𝑦(𝑥) = [𝑥 + 𝑥 + 2] is the required polynomial.

76
6. Using Newton’s forward interpolation formula, find the cubic polynomial which takes

the following values .

𝑿 0 1 2 3

𝒀 1 2 1 10

Solution:

𝒙 𝒚 ∆𝒚 ∆𝟐 𝒚 ∆𝟑 𝒚

𝑥 0 𝑦 1

2 − 1 = 1(∆𝑦 )

𝑥 1 𝑦 2 −1 − 1

= −2 (∆ 𝑦 )

1 − 2 = −1(∆𝑦 ) 10 + 2

𝑥 2 𝑦 1 = 12 (∆ 𝑦 )

9+1 = 10 (∆ 𝑦 )

10 − 1 = 9(∆𝑦 )

𝑥 3 𝑦 10

We will use forward difference formula

𝑢(𝑢 − 1) 𝑢(𝑢 − 1)(𝑢 − 2)

𝑦(𝑥) = 𝑦 + 𝑢∆𝑦 + ∆ 𝑦 + ∆ 𝑦 +⋯
2! 3!

Where 𝑢 = = =𝑥⇒𝑢=𝑥

𝑥(𝑥 − 1) 𝑥(𝑥 − 1)(𝑥 − 2)

∴ 𝑦(𝑥) = 1 + 𝑥(1) + (−2) + (12)
2! 3!

𝑥 −𝑥 𝑥(𝑥 − 1)(𝑥 − 2)
=1+𝑥− (2) + (12)
2 6

77
 = 1 + 𝑥 − 𝑥 + 𝑥 + 2𝑥(𝑥 − 1)(𝑥 − 2)

= 1 + 𝑥 − 𝑥 + 𝑥 + 2𝑥(𝑥 − 3𝑥 + 2)

= 1 + 𝑥 − 𝑥 + 𝑥 + 2𝑥 − 6𝑥 + 4𝑥

= 1 + 6𝑥 − 7𝑥 + 2𝑥

∴ 𝑦(𝑥) = 2𝑥 − 7𝑥 + 6𝑥 + 1

𝑦(4) = 𝑃 (4) = 2(4) − 7(4) + 6(4) + 1

= 2(64) − 7(16) + 24 + 1

= 41

7. From the given table compute the value of sin 𝟑𝟖°

𝒙 0 10 20 30 40

𝒔𝒊𝒏 𝒙 0 0.17365 0.34202 0.5 0.64279

Solution:

We form the difference table:

𝑥 𝑌 = 𝑓(𝑥) ∆𝑦 ∆ 𝑦 ∆ 𝑦 ∆ 𝑦

(𝑦 )

0 0 (∆𝑦 )

0.17365 (∆ 𝑦 )

10 0.17365 -0.00528 (∆ 𝑦 )

0.16837 -0.00511 (∆ 𝑦 )

20 0.34202 -0.01039 0.00031

0.15798 -0.00487 (∇ 𝑦 )

78
 30 0.5 -0.01519 (∇ 𝑦 )

0.14279 (∇ 𝑦 )

40 0.64279

(𝑥 ) (𝑦 )

We will use backward difference formula

𝑣(𝑣 + 1) 𝑣(𝑣 + 1)(𝑣 + 2) 𝑣(𝑣 + 1)(𝑣 + 2)(𝑣 + 3)

𝑦(𝑥) = 𝑦 + 𝑣∇𝑦 + ∇ 𝑦 + ∇ 𝑦 + ∇ 𝑦 +⋯
2! 3! 4!

.
Where 𝑣 = = = −0.2

𝑦(38°) = 0.64279 − 0.028 − 0.0127 + 0.0290 𝑦(38°) = 0.64249

sin 38° = 0.61568

DIFFERENTION USING INTERPOLATION FORMULA

dx d2y
8. Construct and 2 at x  51, from the following data:
dy d x

X: 50 60 70 80 90

Y: 19.96 36.65 58.81 77.21 94.61

Solution:

Given x= 51, x0 = 50 h = 60 – 50 = 10

x  x0 51  50
u   0.1
h 10

At x = 51, u = 0.1

Difference table

x y = f(x) y 2 y 3 y 4 y

79
 50 19.96

16.69

60 36.65 5.47

22.16 -9.23

70 58.81 -3.76 11.99

18.40 2.76

80 77.21 -1.00

17.40

90 94.61

W.K.T the Newton’s forward difference formula is

 dy   dy  1 (2u  1) 2 (3u 2  6u  2) 3
f '( x)       [ y 0   y0   y0
 dx  x  x0  dx u 0.1 h 2! 3!

(4u 3  18u 2  22u  6) 4

  y0  .....]
3!

 dy  1 (0.2  1)  (3(0.1) 2 )  6(0.1)  2 

f '(51)    = [16.69  (5.47) +   (9.23) 
 dx u 0.1 10 2  3! 

 (4(0.1)3  18(0.1) 2 )  22(0.1)  6 

  (11.99)  ......]
 24 

1
= 16.69  2.188  2.1998  1.9863
10

f '(51)  1.0316

 d2y  1 (6u 2  18u  11) 4

f "( x)   2   2 [ 2 y0  (u  1) 3 y0   y0  .....]
 dx u 0.1 h 12

1  (6(0.1) 2  18(0.1)  11) 

f "(51)   5.47  (0.1  1)( 9.23)  (11.99) 
100  12 

80
 1
= 5.47  8.307  9.2523
100

f "(51)  0.2303

9. For the given data, find the first two derivative at x=1.1

x 1.0 1.1 1.2 1.3 1.4 1.5 1.6

y 7.989 8.403 8.781 9.129 9.451 9.750 10.031

Solution:

The difference table is as follows

X y=f(x) ∆𝑦 ∆ 𝑦 ∆ 𝑦 ∆ 𝑦 ∆ 𝑦 ∆ 𝑦

1 7.989

0.414

1.1 8.403 -0.036

0.378 0.006

1.2 8.781 -0.030 -0.002

0.348 0.004 0.001

1.3 9.129 -0.026 -0.001 0.002

0.322 0.003 0.003

1.4 9.451 -0.023 0.002

0.299 0.005

1.5 9.750 -0.018 0

0.281

1.6 10.031

𝑑𝑦 1 (2𝑢 − 1) (3𝑢 − 6𝑢 + 2) 𝑥−𝑥

𝑓 ′ (𝑥) = = ∆𝑦 + ∆ 𝑦 + ∆ 𝑦 +⋯ 𝑊ℎ𝑒𝑟𝑒 𝑢 =
𝑑𝑥 ℎ 2! 3! ℎ

81
 𝑥−𝑥 1
𝑢= = =1
ℎ 1

𝑑𝑦 1 (2(1) − 1) (3(1 ) − 6(1) + 2)

= 0.414 + (−0.036) + (0.006) + ⋯
𝑑𝑥 . 1 2! 3!

𝑑𝑦
= 0.3950
𝑑𝑥 .

NUMERICAL INTEGRATION BY TRAPEZOIDAL METHOD

𝟔 𝒅𝒙
10. Evaluate ∫𝟎 by i) Trapezoidal rule ii) Simpson’s rule. And compare the result with
𝟏 𝒙𝟐

its actual integration value.

Solution:

Here 𝑦(𝑥) =

Let ℎ = 1

x :0 1 2 3 4 5 6

y: 1 0.5 0.2 0.1 0.058824 0.038462 0.27027

We know that for Trapezoidal rule

𝑑𝑥 ℎ
= [(𝑦 + 𝑦 ) + 2(𝑦 + 𝑦 + 𝑦 + 𝑦 + 𝑦 )]
1+𝑥 2

𝑑𝑥 1
= [(1 + 0.27027) + 2(0.5 + 0.2 + 0.1 + 0.058824 + 0.038462)]
1+𝑥 2

𝑑𝑥
= 1.41079950
1+𝑥
82
 We know that Simpson’s one third rule is

𝑑𝑥 ℎ
= [(𝑦 + 𝑦 ) + 4(𝑦 + 𝑦 + 𝑦 ) + 2(𝑦 + 𝑦 )]
1+𝑥 3

𝑑𝑥 1
= [(0.5 + 0.027027) + 4(0.5 + 0.1 + 0.038462) + 2(0.2 + 0.58824)]
1+𝑥 3

𝑑𝑥
= 1.28241
1+𝑥

We know that Simpson’s three – eight rule is

3ℎ
𝑓(𝑥) 𝑑𝑥 = [(𝑦 + 𝑦 ) + 3(𝑦 + 𝑦 + 𝑦 + 𝑦 ) + 2(𝑦 )]
8

𝑑𝑥 3
= [(1 + 0.027027) + 3(0.5 + 0.2 + 0.058824 + 0.038462) + 2(0.1)]
1+𝑥 8

𝑑𝑥
= 1.35708188
1+𝑥

By actual integration,

𝑑𝑥
= (𝑡𝑎𝑛 𝑥) = 𝑡𝑎𝑛 6 = 1.40564764
1+𝑥

Conclusion:

Here the value by trapezoidal rule is closer to the actual value than the value by Simpson’s

rule.

𝟏.𝟑
11. Take 𝒉 = 𝟎. 𝟎𝟓 , evaluate ∫𝟏 √𝒙 𝒅𝒙 using Trapezoidal rule and Simpson’s three-eighth

rule. Solution:

83
 x 1 1.05 1.1 1.15 1.2 1.25 1.3

y 1 1.0247 1.0488 1.0724 1.0954 1.118 1.1402

We know that for Trapezoidal rule

𝟏.𝟑
ℎ
√𝒙 𝒅𝒙 = [(𝑦 + 𝑦 ) + 2(𝑦 + 𝑦 + 𝑦 + 𝑦 + 𝑦 )]
2
𝟏

𝟏.𝟑
0.05
√𝒙 𝒅𝒙 = [(1 + 1.1402) + 2(1.0247 + 1.0488 + 1.0724 + 1.0954 + 1.118)]
2
𝟏

𝟏.𝟑

√𝒙 𝒅𝒙 = 0.3215
𝟏

We know that Simpson’s three – eight rule is

3ℎ
𝑓(𝑥) 𝑑𝑥 = [(𝑦 + 𝑦 ) + 3(𝑦 + 𝑦 + 𝑦 + 𝑦 ) + 2(𝑦 )]
8

𝟏.𝟑
3(0.05)
√𝒙 𝒅𝒙 = [(1 + 1.1402) + 3(1.0247 + 1.0488 + 1.0954 + 1.118) + 2(1.0724)]
8
𝟏

𝟏.𝟑

√𝒙 𝒅𝒙 = 0.3215
𝟏

DOUBLE INTEGRALS USING TRAPEZOIDAL AND SIMPSON’S RULES

𝟏.𝟐 𝟏.𝟒 𝟏
12. Evaluate ∫𝟏 ∫𝟏 𝒅𝒙 𝒅𝒚 by using Trapezoidal rule taking h=0.1 and k=0.1
𝒙 𝒚

Solution:

84
 y\x 1 1.1 1.2 1.3 1.4

1 0.5000 0.4762 0.4545 0.4348 0.4167

1.1 0.4762 0.4545 0.4348 0.4167 0.4000

1.2 0.4545 0.4348 0.4167 0.4000 0.3846

I= [(sum of values of f at the four corners)

+ 2 (sum of values of f at the remaining nodes on the boundary)

+ 4(sum of the values of f at the interior nodes)]

(0.1)(0.1)
I = [(0.5000 + 0.4167 + 0.3846 + 0.4545)
4

+ 2(0.4762 + 0.4545 + 0.4348 + 0.4000 + 0.4000 + 0.4167 + 0.4348 + 0.4762)

+ 4(0.4545 + 0.4348 + 0.4167)]

𝐼 = 0.0349

𝟏 𝟏 𝟏
13. Evaluate ∫𝟎 ∫𝟎 𝒅𝒙 𝒅𝒚 by using Trapezoidal rule taking h=0.5 and k=0.25
𝒙 𝒚 𝟏

Solution:

0 0.5 1

0 1 0.6667 0.5

0.25 0.8 0.5714 0.4444

0.5 0.6667 0.5 0.40

0.75 0.5714 0.4444 0.3636

1 0.50 .40 0.3333

I= [(sum of values of f at the four corners)

+ 2 (sum of values of f at the remaining nodes on the boundary)

+ 4(sum of the values of f at the interior nodes)]

85
 ( . )( . )
I= [(1 + 0.5 + 0.3333 + 0.5) + 2(0.667 + 0.4444 + 0.40 + 0.3636 + 0.40 + 0.5714 +

0.6667 + 0.8) + 4(0.5714 + 0.5 + 0.4444)]

= 0.5319

𝟑 𝟐 𝟏
14. Evaluate ∫𝟏 ∫𝟏 𝒅𝒙 𝒅𝒚 by using Trapezoidal rule taking h=0.5 and k=0.5
𝒙𝒚

Solution:

1 1.5 2

1 1 0.667 0.5

1.5 0.667 0.4444 0.3333

2 0.5 0.3333 0.25

2.5 0.4000 0.2667 0.2000

3 0.3333 4.5000 0.1667

I= [(sum of values of f at the four corners)

+ 2 (sum of values of f at the remaining nodes on the boundary)

+ 4(sum of the values of f at the interior nodes)]

( . )( . )
I= [(1 + 0.5 + 0.3333 + 0.1667) + 2(0.667 + 0.3333 + 0.25 + 0.2 + 4.5 + 0.4 +

0.5 + 0.667) + 4(0.4444 + 0.3333 + 0.2667)]

𝐼 = 1.3258

𝟏/𝟐 𝟏/𝟐 𝒔𝒊𝒏(𝒙𝒚)

15. Evaluate ∫𝟎 ∫𝟎 𝒅𝒙 𝒅𝒚 using Simpson’s rule with h=1/4=k
𝟏 𝒙𝒚

Solution:

( )
Let 𝑓(𝑥, 𝑦) =

The values of 𝑓(𝑥, 𝑦) at the nodal points are given in the following table
86
 0 1/4 ½

0 0 0 0

¼ 0 0.0588 0.1108

1/2 0 0.1108 0.1979

By Simpson’s rule, I = [(sum of values of f at the four corners)

+ 2 (sum of the values of f at the odd position on the boundary except the corners)

+ 4 (sum of the values of f at the even position on the boundary)

+ {4 (sum of the values of f at odd positions) + 8 (sum of the values of

f at even positions) on the odd row of the matrix except boundary rows}

+ {8 (sum of the values of f at the odd positions)+16 (sum of the

Values of f at the even position) on the even rows of the matrix}]

I = [(0 + 0 + 0.1979 + 0) + 4(0 + 0 + 0.1108 + 0.1108) + 16(0.0588)]

9

𝐼 = 0.0141

𝟏.𝟒 𝟐.𝟒 𝟏
16. Evaluate ∫𝟏 ∫𝟐 𝒅𝒙 𝒅𝒚 by using Trapezoidal rule taking h=0.1 and k=0.1
𝒙𝒚

and verify with actual integration .

Solution:

y\x 1 1.1 1.2 1.3 1.4

1 0.5 0.4762 0.4545 0.4348 0.4167

1.1 0.4545 0.4329 0.4132 0.3953 0.3788

1.2 0.4167 0.3968 0.3788 0.3623 0.3472

1.3 0.3846 0.3663 0.3497 0.3344 0.3205

87
 1.4 0.3571 0.3401 0.3247 0.3106 0.2976

I= [(sum of values of f at the four corners)

+ 2 (sum of values of f at the remaining nodes on the boundary)

+ 4(sum of the values of f at the interior nodes)]

(0.1)(0.1)
I = [(0.5000 + 0.4167 + 0.3571 + 0.2976)
4

+ 2(0.3846 + 0.4167 + 0.4545 + 0.4762 + 0.4545 + 0.4348 + 0.3788 + 0.3472

+ 0.3205 + 0.3106 + 0.3247 + 0.3401)

+ 4(0.4329 + 0.4132 + 0.3953 + 0.3968 + 0.3788 + 0.3623 + 0.3663 + 0.3497

+ 0.3344)]

𝐼 = 0.0614

By actual integration:

. . . .
1 1 1
𝑑𝑥 𝑑𝑦 = 𝑑𝑦) ( 𝑑𝑥
𝑥𝑦 𝑦 𝑥

. .
= (𝑙𝑜𝑔 𝑦) (𝑙𝑜𝑔 𝑦)

= (𝑙𝑜𝑔 1.4)[𝑙𝑜𝑔2.4 − 𝑙𝑜𝑔2]

= 𝑙𝑜𝑔(1.4)𝑙𝑜𝑔(1.2)

. .
1
𝑑𝑥 𝑑𝑦 = 0.0613
𝑥𝑦

UNIT-V

NUMERICAL SOLUTION OF ORDINARY DIFFERENTIAL EQUATIONS

88
 PART-A

TAYLOR SERIES METHOD

1. Using Taylor series method fine 𝒚(𝟏. 𝟏) given that 𝒚′ = 𝒙 + 𝒚, 𝒚(𝟏) = 𝟎

Solution:
Given 𝑦 ′ = 𝑥 + 𝑦 and 𝑥 = 1, 𝑦 = 0
We know that Taylor series formula is
(𝑥 − 𝑥 ) ′ (𝑥 − 𝑥 ) ′′ (𝑥 − 𝑥 ) ′′′
𝑦 = 𝑦 + 𝑦 + 𝑦 + 𝑦 +⋯
1! 2! 3!
𝑦′ = 𝑥 + 𝑦 𝑦′ = 1 + 0 = 1
𝑦 ′′ = 1 + 𝑦 ′ 𝑦 ′′
=1+1=2
𝑦 ′′′ = 𝑦 ′′ 𝑦 ′′′
=2
(𝑥 − 1) (𝑥 − 1)
𝑦 = 0 + (𝑥 − 1) + (2) + (2)
2 6
(1.1 − 1) (1.1 − 1)
𝑦(1.1) = 0 + (1.1 − 1) + (2) + (2)
2 6
𝑦 = 𝑦(1.1) = 0.1103
𝒅𝒚
2. Find 𝒚(𝟎. 𝟏) if = 𝟏 + 𝒚, 𝒚(𝟎) = 𝟏 using Taylor series method.
𝒅𝒙

Solution:

Given 𝑦 ′ = 1 + 𝑦 and 𝑥 = 0, 𝑦 = 1

We know that Taylor series formula is

(𝑥 − 𝑥 ) ′ (𝑥 − 𝑥 ) ′′ (𝑥 − 𝑥 ) ′′′
𝑦 = 𝑦 + 𝑦 + 𝑦 + 𝑦 +⋯
1! 2! 3!

𝑦′ = 1 + 𝑦 𝑦′ = 1 + 1 = 2

𝑦 ′′ = 𝑦 ′ 𝑦 ′′
=2

𝑦 ′′′ = 𝑦 ′′ 𝑦 ′′′
=2

(𝑥 − 0) (𝑥 − 0) (𝑥 − 0)
𝑦 = 1 + (𝑥 − 0)2 + (2) + (2) + (2)
2 6 24

𝑥 𝑥
= 1 + 2𝑥 + 𝑥 + +
3 12

(0.1) (0.4)
𝑦(0.1) = 1 + 2(0.1) + (0.1) + +
3 12

89
 𝑦 = 𝑦(0.1) = 1.2103

3. State the advantages and disadvantages of the Taylor’s series method.

Solution:

The method gives a straight forward adaptation of classic calculus to develop the solution as
an infinite series. It is a powerful single step method if we are able to find the successive
derivatives easily.
If f(x,y) involves some complicated algebraic structures then the calculation of higher
derivatives becomes tedious and the method fails.

EULER AND MODIFIED EULER METHOD

𝒅𝒚
4. State Euler’s method to solve = 𝒇(𝒙, 𝒚)𝒘𝒊𝒕𝒉 𝒚(𝒙𝟎 ) = 𝒚𝟎 .
𝒅𝒙

Solution:

𝑦 = 𝑦 + ℎ𝑓(𝑥 , 𝑦 ) where 𝑛 = 0,1,2 …

𝒅𝒚
5. State Modified Euler’s method to solve = 𝒇(𝒙, 𝒚)𝒘𝒊𝒕𝒉 𝒚(𝒙𝟎 ) = 𝒚𝟎 .
𝒅𝒙

Solution:

 h h 
y1  y0  hf  x0  , y0  f ( x0 , y0 ) 
 2 2 
𝒅𝒚
6. Find 𝒚(𝟎. 𝟏)by using Euler’s method given that = 𝒙 + 𝒚, 𝒚(𝟎) = 𝟏.
𝒅𝒙

Solution:

Given 𝑦 ′ = 𝑥 + 𝑦, 𝑥 = 0, 𝑦 = 1
By Euler’s method
𝑦 = 𝑦 + ℎ𝑓(𝑥 , 𝑦 )
𝑦 = 1 + (0.1)(0 + 1) = 1 + 0.1 = 1.2
𝑦 = 𝑦(0.1) = 1.2

7. Find 𝒚(𝟎. 𝟐) for the equation 𝒚′ = 𝒚 + 𝒆𝒙 , given that 𝒚(𝟎) = 𝟎 by using Euler’s method.

Solution:

Given 𝑦 ′ = 𝑦 + 𝑒 , 𝑥 = 0, 𝑦 = 0, ℎ = 0.2
90
 By Euler algorithm, 𝑦 = 𝑦 + ℎ𝑓(𝑥 , 𝑦 )
= 0 + 0.2𝑓(0,0)
= 0.2[0 + 𝑒 ] = 0.2
𝑦(0.2) = 0.2

RUNGE-KUTTA METHOD FOR SOLVING FIRST AND SECOND ORDER EQUATIONS

8. State the fourth order Runge-Kutta algorithm.

Solution:

Let h denote the interval between equidistant values of x. if the initial values are (𝑥 , 𝑦 ),
the first increment in y is computed from the formulas.
𝑘 = ℎ𝑓(𝑥 , 𝑦 )
ℎ 𝑘
𝑘 = ℎ𝑓 𝑥 + , 𝑦 +
2 2
ℎ 𝑘
𝑘 = ℎ𝑓 𝑥 + , 𝑦 +
2 2
𝑘 = ℎ𝑓(𝑥 + ℎ, 𝑦 + 𝑘 )
1
∆𝑦 = (𝑘 + 2𝑘 + 2𝑘 + 𝑘 )
6
𝑥 = 𝑥 + ℎ, 𝑦 = 𝑦 + ∆𝑦

The increment in y in the second interval is computed in a similar manner using the
same four formulas, using the values 𝑥 , 𝑦 in the place of 𝑥 , 𝑦 respectively.
𝑓 (𝑥, 𝑦, 𝑧) = 𝑧
𝑓 (𝑥, 𝑦, 𝑧) = −𝑥𝑧 − 𝑦
By Runge- Kutta method
𝑘 = ℎ𝑓 (𝑥 , 𝑦 , 𝑧 ) 𝑙 = ℎ𝑓 (𝑥 , 𝑦 , 𝑧 )
= (0.1)𝑓 (0,1,0) = (0.1)𝑓 (0,1,0)
= (0.1)(0) = (0.1)(0 − 1)
𝑘 =0 𝑙 = −0.1

MILNE’S PREDICTOR AND CORRECTOR METHODS

9. State Milne’s predictor-corrector formula.

Solution:

Milne’s Predictor Formula:

91
 4ℎ
𝑦 , =𝑦 + (2𝑦 ′ − 𝑦′ + 2𝑦 ′ )
3

Milne’s Corrector Formula:

ℎ
𝑦 , =𝑦 + (2𝑦 ′ − 4𝑦 ′ + 𝑦 ′ )
3

10. Distinguish between single step methods and multi-step methods.

Solution:
single step method multi-step method
Taylor’s series, Euler’s, Modified Euler’s, Milne’s and Adams predictor - corrector
Runge – Kutta method of fourth order method
One prior value is required for finding the Four prior value are required for finding
value of 𝑦 at 𝑥 the value of 𝑦 at 𝑥
11. What are multi-step methods? How are they better than single step methods?
Solution:
One step method: We use data of just one proceeding step.
Multi step method: We use data from more than one of the proceeding steps.

PART-B

TAYLOR SERIES METHOD

𝒅𝒚
1. Find the value of 𝒚 at 𝒙 = 𝟎. 𝟏, 𝟎. 𝟐 given that = 𝒙𝟐 𝒚 − 𝟏, 𝒚(𝟎) = 𝟏,by Tailor’s series method
𝒅𝒙

up to four terms.
Solution:
Given 𝑦 ′ = 𝑥 𝑦 − 1 and 𝑥 = 0, 𝑦 = 1
92
 We know that Taylor series formula is

(𝒙 𝒙𝟎 ) ′ (𝒙 𝒙𝟎 )𝟐 ′′ (𝒙 𝒙𝟎 )𝟑 ′′′
𝒚 = 𝒚𝟎 + 𝒚𝟎 + 𝒚𝟎 + 𝒚𝟎 +⋯ … (1)
𝟏! 𝟐! 𝟑!

𝑦′ = 𝑥 𝑦 − 1 𝑦 ′ = 0 − 1 = −1

𝑦 ′′ = 2𝑥𝑦 + 𝑥 𝑦 ′ 𝑦 ′′ = 2(0)(1) + 0(−1) = 0

𝑦 ′′′ = 2 𝑥𝑦 ′ + 𝑦 + 𝑥 𝑦 ′′ + 𝑦 ′ 2𝑥 𝑦 ′′′ = 2(1) + 4(0)(−1) + 0 = 2

= 2𝑦 + 4𝑥𝑦 ′ + 𝑥 𝑦 ′′

𝑦 = 2𝑦 ′ + 4 𝑥𝑦 ′′ + 𝑦 ′ + 𝑥 𝑦 ′′′ + 𝑦 ′′′ 2𝑥 𝑦 = 6(−1) + 6(0)(0) + (0)(2)

= 6𝑦 ′ + 6𝑥𝑦 ′′ + 𝑥 𝑦 ′′′ = −6

Substituting in equation (1) we get

( ) ( ) ( )
𝑦 = 1 + (𝑥 − 0)(−1) + (0) + (2) + (−6)

𝑥 𝑥
𝑦 =1−𝑥+ −
3 4

To find y (0.1)

0.1 0.1
𝑦(0.1) = 1 − 0.1 + −
3 4

𝑦(0.1) = 1 − 0.1 + 0.00033 − 0.000025

𝑦(0.1) = 0.900305

To find y (0.2)

0.2 0.2
𝑦(0.2) = 1 − 0.2 + −
3 4

𝑦(0.2) = 1 − 0.2 + 0.0026 + −0.0004

𝑦(0.2) = 0.8022

𝑥 = 0.1 , 𝑦 = 00.0993, ℎ = 0.1

(0.1) (0.1)
𝑦 = 𝑦(0.2) = 0.09933 + (0.1)(0.9801334) + (−0.3946868) + (−3.84159)
2 6

𝑦(0.2) = 0.19467

93
2. Determine the value of 𝒚(𝟎. 𝟒) using milnes’s method given 𝒚′ = 𝒙𝒚 + 𝒚𝟐 ,𝒚(𝟎) = 𝟏 . Using
Taylor series method obtain the values of 𝒚(𝟎. 𝟏) and y(0.2) and y(0.3).

Solution :

Given 𝑦 ′ = 𝑥𝑦 + 𝑦 and 𝑥 = 0, 𝑦 = 1,
By Taylor series formula is
(𝒙 𝒙𝟎 ) ′ (𝒙 𝒙𝟎 )𝟐 ′′ (𝒙 𝒙𝟎 )𝟑 ′′′
𝒚 = 𝒚𝟎 + 𝒚𝟎 + 𝒚𝟎 + 𝒚𝟎 +⋯ … (1)
𝟏! 𝟐! 𝟑!

𝑦 ′ = 𝑥𝑦 + 𝑦 𝑦′ = 1

𝑦 ′′ = 𝑥𝑦 ′ + 𝑦 + 2 𝑦 ′ 𝑦 ′′ = 1 + 2(1)(1) = 1

𝑦 ′′′ = 𝑥𝑦 ′ + 𝑦 ′ + 𝑦 ′ + 2𝑦𝑦 ′′ + 2𝑦 ′ 𝑦 ′ 𝑦 ′′′ = 2 + 6 + 2 = 10

= 𝑥𝑦 ′ + 2𝑦 ′ + 2𝑦𝑦 ′′
+ 2𝑦 ′
𝑦 = 𝑥𝑦 ′′′ + 𝑦 ′′ + 2𝑦 ′′ + 2𝑦 ′ 𝑦 ′′ 𝑦 = 9 + 12 + 20 = 41
+ 2𝑦 𝑦 ′′′ + 4 𝑦 ′ 𝑦 ′′
= 𝑥𝑦 ′′′ + 3𝑦 ′′ + 4 𝑦 ′ 𝑦 ′′ +2𝑦 𝑦 ′′′
Substituting in equation (1) we get
(0.1) (0.1) (0.1)
𝑦 = 𝑦(0.1) = 1 + 0.1(1) + (3) + (10) + (41)
2 6 24
𝑦(0.1) = 1.11684

(0.2) (0.2) (0.2)

𝑦 = 𝑦(0.2) = 1 + 0.2(1) + (3) + (10) + (41)
2 6 24
𝑦(0.2) = 1.276067

(0.3) (0.3) (0. )

𝑦 = 𝑦(0.3) = 1 + 0.3(1) + (3) + (10) + (41)
2 6 24
𝑦(0.1) = 1.48384

X 0 0.1 0.2 0.3

Y 1 1.11684 1.27607 1.49384

𝟒𝒉
𝒚𝟒, 𝒑 = 𝒚𝟎 + 𝟐𝒚′𝟏 − 𝒚′𝟐 + 𝟐𝒚′𝟑
𝟑
4(0.1)
𝑦 , =1+ [2(1.35902) − 1.88357 + 2(2.67974)]
3
𝑦 , = 1.82586
𝑦 ′ = (0.4)1.82586 + 1.82586 = 4.06411
By Mile’s corrector formula is
𝒉 ′
𝒚𝟒, 𝒄 = 𝒚𝟐 + 𝒚 + 𝟒𝒚′𝟑 + 𝒚′𝟒
𝟑 𝟐
94
 0.1
= 1.27607 + [1.88357 + (2.67974) + 4.06411]
3
𝒚𝟒, 𝒄 = 𝟏. 𝟖𝟑𝟎𝟗𝟔

𝑦 = 1.83096

𝒅𝒚
3. Using Taylor series method fin y at x=1.1 by solving the equation if = 𝒙𝟐 + 𝒚𝟐 , 𝒚(𝟏) = 𝟐.
𝒅𝒙
Carryout the computations upto fourth order derivative.

Solution:

Given initial condition 𝑥 = 1, 𝑦 = 2, ℎ = 0.1

We know that Taylor series formula is
( ) ( ) ( )
𝑦=𝑦 + 𝑦′ + 𝑦 ′′ + 𝑦 ′′′ + ⋯ … (1)
! ! !
𝑦 ′ = 𝒙𝟐 + 𝒚 𝟐
𝑦′ = 1 + 2 = 5

𝑦 ′′ = 2𝑥 + 2𝑦𝑦′ 𝑦 ′′ = 2(1) + 2(2)(5) = 22

𝑦 ′′′ = 2 + 2𝑦𝑦" + 2𝑦′ 𝑦 ′′′ = 2 + 2(2)(22) + 2(5) = 140

𝑦 = 2 𝑦𝑦 ′′′ + 2𝑦′𝑦" + 4𝑦′𝑦" 𝑦 = 2(2)(140) + 2(5)(22) + 4(5)(22) = 1220

Substituting in equation (1) we get

(𝑥 − 𝑥 ) (𝑥 − 𝑥 ) (𝑥 − 𝑥 ) (𝑥 − 𝑥 )
𝑦 =2+ (5) + (22) + (140) + (1220) + ⋯
1! 2! 3! 4!
(1.1 − 1) (1.1 − 1) (1.1 − 1) (1.1 − 1)
𝑦 =2+ (5) + (22) + (140) + (1220) + ⋯
1! 2! 3! 4!
(0.1) (0.1) (0.1)
𝑦(1.1) = 2 + 0.1(5) + (22) + (140) + (1220) + ⋯
2 6 24
𝑦 = 2 + 0.5 + 0.11 + 0.023 + 0.00508 = 2.63808

EULER AND MODIFIED EULER METHOD

𝒅𝒚
4. Apply Modified Euler’s method to find 𝒚(𝟎. 𝟐) and 𝒚(𝟎. 𝟒) given that = 𝒙𝟐 + 𝒚𝟐 , 𝒚(𝟎) = 𝟏 by
𝒅𝒙

taking h=0.2

Solution:

Initial conditions are

𝑥 = 0, 𝑦 = 1, ℎ = 0.2

By Euler algorithm

95
 ℎ ℎ
𝑦 = 𝑦 + ℎ𝑓(𝑥 + , 𝑦 + (𝑥 , 𝑦 ′)
2 2

𝐿𝑒𝑡 𝑛 = 0

ℎ 1
𝑦 = 𝑦 + ℎ𝑓(𝑥 + , 𝑦 + (𝑥 , 𝑦 ′)
2 2
0.2 0.2 𝟐
= 1 + (0.2)𝑓(0 + ,1 + (𝟎 + 𝟏𝟐 ))
2 2

= 1 + (0.2)𝑓(0.1,1.1)

= 1 + (0.2) (𝟎. 𝟏)𝟐 + (𝟏. 𝟏)𝟐

= 1 + (0.2)(1.22)

= 1.244

𝑦 = 1.244

𝒚𝟏 = 𝒚(𝟎. 𝟐) = 𝟏. 𝟐𝟒𝟒

𝐿𝑒𝑡 𝑛 = 1,

𝑥 = 0.2, 𝑦 = 1.244, ℎ = 0.2

ℎ ℎ
𝑦 = 𝑦 + ℎ𝑓(𝑥 + , 𝑦 + (𝑥 , 𝑦 ′)
2 2
0.2 0.2
= 1.244 + (0.2)𝑓(0.2 + , 1.244 + ((𝟎. 𝟐)𝟐 + (𝟏. 𝟐𝟒𝟒)𝟐 ))
2 2

= 1.244 + (0.2)𝑓(0.3,1.4028)

= 1.244 + (0.2) (𝟎. 𝟑)𝟐 + (𝟏. 𝟑𝟔𝟖𝟒)𝟐

𝑦 = 1.6556

𝒚𝟐 = 𝒚(𝟎. 𝟒) = 𝟏. 𝟔𝟓𝟓𝟔

dy
5. Evaluate y at x  0.2 given  y  x 2  1, y (0)  0.5 using modified Euler’s method.
dx

Solution:

𝑑𝑦
= y − 𝑥 + 1, 𝑥 = 0, 𝑦 = 0.5 , ℎ = 0.2
𝑑𝑥

By Euler algorithm
ℎ 1
𝑦 = 𝑦 + ℎ𝑓(𝑥 + , 𝑦 + ℎ(𝑥 , 𝑦 ′)
2 2
96
 𝐿𝑒𝑡𝑛 = 0
ℎ 1
𝑦 = 𝑦 + ℎ𝑓(𝑥 + , 𝑦 + ℎ(𝑥 , 𝑦 ′)
2 2
0.2 0.2
= 0.5 + (0.2)𝑓 0 + , 0.5 + (0,0.5)
2 2
𝑓(𝑥 , 𝑦 ) = 𝑦 − 𝑥 + 1, f(0,0.5) = 0.5 + 0 + 1 = 1.5

= 0.5 + (0.2)𝑓[(0.1, 0.5 + 0.1(1.5)]

= 0.5 + (0.2)𝑓(0.1,0.65)
𝑓(0.1,0.65) = 0.65 + (0.1) + 1 = 0.65 − 0.01 + 1
= 1.65 − 0.01 = 1.64
𝑦 = 0.5 + (0.2)(1.64)
0.5 + 0.328 = 0.828

𝒚(𝟎. 𝟐) = 𝟎. 𝟖𝟐𝟖

𝒅𝒚
6. Apply Modified Euler’s method to find 𝒚(𝟎. 𝟏) and 𝒚(𝟎. 𝟐) given that = 𝒙𝟐 + 𝒚𝟐 , 𝒚(𝟎) = 𝟏
𝒅𝒙

Solution:

Initial conditions are

𝑥 = 0, 𝑦 = 1, ℎ = 0.1

By Euler algorithm

ℎ 1
𝑦 = 𝑦 + ℎ𝑓(𝑥 + , 𝑦 + ℎ(𝑥 , 𝑦 ′)
2 2

𝐿𝑒𝑡 𝑛 = 0

ℎ 1
𝑦 = 𝑦 + ℎ𝑓(𝑥 + , 𝑦 + ℎ(𝑥 , 𝑦 ′)
2 2
0.1 0.1
= 1 + (0.1)𝑓(0 + ,1 + (𝟏 + 𝟎))
2 2

= 1 + (0.1)𝑓(0.05,1.05)

= 1 + (0.1) (𝟎. 𝟎𝟓)𝟐 + (𝟏. 𝟎𝟓)𝟐

= 1 + (0.1)(1.105)

= 1.1105

97
 𝑦 = 1.1105

𝒚𝟏 = 𝒚(𝟎. 𝟏) = 𝟏. 𝟏𝟏𝟎𝟓

𝐿𝑒𝑡 𝑛 = 1,

𝑥 = 0.1, 𝑦 = 1.1105, ℎ = 0.1

ℎ 1
𝑦 = 𝑦 + ℎ𝑓(𝑥 + , 𝑦 + ℎ(𝑥 , 𝑦 ′)
2 2
0.1 0.1
= 1.1105 + (0.1)𝑓(0.1 + , 1.1105 + ((𝟎. 𝟐)𝟐 + (𝟏. 𝟏𝟏𝟎𝟓)𝟐 ))
2 2

= 1.1105 + (0.1)𝑓(0.15,1.27321)

= 1.1105 + 0.1((𝟎. 𝟏𝟓)𝟐 + (𝟏. 𝟐𝟕𝟑𝟐𝟏)𝟐 ))

𝑦 = 1.2749

𝒚𝟐 = 𝒚(𝟎. 𝟐) = 𝟏. 𝟐𝟕𝟒𝟗

RUNGE-KUTTA METHOD FOR SOLVING FIRST AND SECOND ORDER EQUATIONS

𝒅𝒚
7. Use Runge-Kutta method of order 4 to find y(1.1) given = 𝒚𝟐 + 𝒙𝒚, 𝒚(𝟏) = 𝟏,
𝒅𝒙

Solution:
𝒅𝒚
Given = 𝒚𝟐 + 𝒙𝒚 , 𝑥 = 1, 𝑦 = 1and ℎ = 0. 1
𝒅𝒙

By Runge-kutta method
𝒌𝟏 = 𝒉𝒇(𝒙𝟎 , 𝒚𝟎 )
= (0.1)𝑓(1,1)
= (0.1)(11 + 1)
𝑘 = 0.2
𝒉 𝒌𝟏
𝒌𝟐 = 𝒉𝒇 𝒙𝟎 + , 𝒚𝟎 +
𝟐 𝟐
0.1 0.2
= (0.1)𝑓 1 + ,1 +
2 2
= (0.1)𝑓(1.05,1.1)
𝑘 = 0.2365
𝒉 𝒌𝟐
𝒌𝟑 = 𝒉𝒇 𝒙𝟎 + , 𝒚𝟎 +
𝟐 𝟐
0.1 0.2365
= (0.1)𝑓 1 + ,1 +
2 2

98
 = (0.1)𝑓(1.05,1.118)

𝑘 = 0.2423

𝒌𝟒 = 𝒉𝒇(𝒙𝟎 + 𝒉, 𝒚𝟎 + 𝒌𝟑 )
= (0.1)𝑓(1 + 0.1,1 + 0.2423)
= (0.1)𝑓(1.1,1.12423)
𝑘 = 0.2909
𝟏
∆𝒚 = (𝒌 + 𝟐𝒌𝟐 + 𝟐𝒌𝟑 + 𝒌𝟒 )
𝟔 𝟏
1
= (0.2 + 2(0.0.2365) + 2(0.2423) + 0.2909)
6
∆𝑦 = 0.2414
𝑦 = 𝑦 + ∆𝑦
= 1 + 0.2414
𝒚(𝟏. 𝟎𝟓) = 𝟏. 𝟐𝟒𝟏𝟒
To find y(1.1):
Here 𝑥 = 1.05, 𝑦 = 1.2414and ℎ = 0.1
𝒌𝟏 = 𝒉𝒇(𝒙𝟏 , 𝒚𝟏 )
= (0.1)𝑓(1.05,1.2414)
= (0.1)(2.84454)
𝑘 = 0.28445
𝒉 𝒌𝟏
𝒌𝟐 = 𝒉𝒇 𝒙𝟏 + , 𝒚𝟏 +
𝟐 𝟐
0.1 0.28445
= (0.1)𝑓 1.05 + , 1.2414 +
2 2
= (0.1)𝑓(1.1,1.3836)
𝑘 = 0.27133
𝒉 𝒌𝟐
𝒌𝟑 = 𝒉𝒇 𝒙𝟏 + , 𝒚𝟏 +
𝟐 𝟐
0.1 0.27133
= (0.1)𝑓 1.05 + , 1.2414 +
2 2
= (0.1)𝑓(1.1,1.37706)
𝑘 = 0.34110
𝒌𝟒 = 𝒉𝒇(𝒙𝟏 + 𝒉, 𝒚𝟏 + 𝒌𝟑 )
= (0.1)𝑓(1.05 + 0.1,1.2421 + 0.34110)
= (0.1)𝑓(1.15,1.5825)
𝑘 = 0.43241

99
 𝟏
∆𝒚 = (𝒌 + 𝟐𝒌𝟐 + 𝟐𝒌𝟑 + 𝒌𝟒 )
𝟔 𝟏
1
= (0.2844 + 2(0.27133) + 2(0.34110) + 0.43241)
6
∆𝑦 = 0.3236016
𝑦 = 𝑦 + ∆𝑦
= 1.2414 + 0.3236016
𝑦(1.1) = 1.565001

MILNE’S PREDICTOR AND CORRECTOR METHODS

8. Use Milne’s predictor – corrector formula to find 𝒚(𝟎. 𝟒)

𝒅𝒚 𝟏 𝒙 𝟐 𝒚𝟐
Given = , 𝒚(𝟎) = 𝟏 , 𝒚(𝟎. 𝟏) = 𝟏. 𝟎𝟔, 𝒚(𝟎. 𝟐) = 𝟏. 𝟏𝟐 𝒂𝒏𝒅 𝒚(𝟎. 𝟑) = 𝟏. 𝟐𝟏
𝒅𝒙 𝟐

Solution:

Given = 𝑦 = (1 + 𝑥 )𝑦 𝑎𝑛𝑑 ℎ = 0.1

𝑥 = 0, 𝑥 = 0.1, 𝑥 = 0.2, 𝑥 = 0.3, 𝑥 = 0.4, 𝑥 = 0.5

𝑦 = 1, 𝑦 = 1.06, 𝑦 = 1.12, 𝑦 = 1.21, 𝑦 =?

Milene’s Predictor formula we have,

𝟒𝒉
𝒚𝒏 𝟏, 𝒑 = 𝒚𝒏 𝟑 + [𝟐𝒚 𝒏 𝟐 −𝒚 𝒏 𝟏
+ 𝟐𝒚′𝒏 … … … … … … … … (1)
𝟑

To get 𝑦 , put 𝑛 = 3 in (1) we get

𝑦 ,𝑝 = 𝑦 + [2𝑦 − 𝑦 + 2𝑦 ] … … … … … … … … … … … … (2)

1
𝑦 = (1 + 𝑥 )𝑦
2

1
= (1 + 𝑥 )𝑦
2

1
= [1 + (0.1) ](1.06)
2

𝑦 = 0.56742… … … … … … … … … .(3)

100
 1
𝑦 = (1 + 𝑥 )𝑦
2

1
= [1 + (0.2) ](1.12)
2

1
= (1 + 0.04)(1.2544)
2

𝑦 = 0.6529 … … … … … … … … … … … … … (4)

1
𝑦 = (1 + 𝑥 )𝑦
2

1
= [1 + (0.3) ](1.21)
2

1
= [1 + 0.09](1.464)
2

𝑦 = 0.79793 … … … … … … … … … … … … (5)

Substituting (3),(4) and (5) in (2) we get,

4(0.1)
𝑦 , =1+ [2(0.56742) − 0.65229 + 2(0.79793)]
2

0.4
=1+ [1.13484 − 0.65229 + 1.56586]
3

= 1 + 0.27712

𝒚(𝟎. 𝟒) = 𝟏. 𝟐𝟕𝟕𝟏𝟐

Milne’s corrector formula we have

𝒉
𝒚𝒏 𝟏 ,𝒄 = 𝒚𝒏 𝟏 + 𝒚 𝒏 𝟏
+ 𝟒𝒚 𝒏
+ 𝒚𝒏 𝟏
𝟑

To get 𝑦 , put 𝑛 = 3 we get

𝑦 , = 𝑦 + (𝑦 + 4𝑦 + 𝑦 ) … … … …… … … …(6)

1
𝑦 = (1 + 𝑥 )𝑦
2

1
= [1 + (0.4) ](1.27712)
2
101
 1
= (1 + 0.16)(1.63104)
2

1
= (1.16)(1.63104)
2

= 0.94600… … … … … … … … … … … … … (7)

Substituting (4), (5), (7) in (6) we get,

0.1
𝑦 , = 1.12 + [0.65229 + 4(0.79793) + 0.94600]
3

0.1
= 1.12 + [4.79001]
3

= 1.12 + 0.159667

𝒚(𝟎. 𝟒) = 𝟏. 𝟐𝟕𝟗𝟔𝟔

9. Using Milne’s predictor and corrector formulae , find 𝒚(𝟒. 𝟒) given

𝟓𝒙𝒚 + 𝒚𝟐 − 𝟐 = 𝟎, 𝒚(𝟒) = 𝟏, 𝒚(𝟒. 𝟏) = 𝟏. 𝟎𝟎𝟒𝟗, 𝒚(𝟒. 𝟐) = 𝟏. 𝟎𝟎𝟗𝟕, 𝒚(𝟒. 𝟑) = 𝟏. 𝟎𝟏𝟒𝟑

Solution:

Given 𝑦 = , 𝑥 = 4, 𝑥 = 4.1, 𝑥 = 4.2, 𝑥 = 4.3, 𝑥 = 4.4

𝑦 = 1, 𝑦 = 1.0049, 𝑦 = 1.0097, 𝑦 = 1.0143

2−𝑦 2 − (1.0049)
𝑦 = = = 0.0493
5𝑥 5(4.1)
2−𝑦 2 − (1.0097)
𝑦 = = = 0.0467
5𝑥 5(4.2)
2−𝑦 2 − (1.0143)
𝑦 = = = 0.0452
5𝑥 5(4.3)
By Mile’s predictor formula is
𝟒𝒉
𝒚𝟒, 𝒑 = 𝒚𝟎 + [𝟐𝒚𝟏 − 𝒚𝟐 + 𝟐𝒚𝟑 ]
𝟑
4(0.1)
𝑦 , =1+ [2(0.0493 − 0.0467 + 2(0.0452)]
3

𝑦 , = 1.01897
2−𝑦 2 − (1.1897)
𝑦 = = = 0.0437
5𝑥 5(4.4)
102
 By Mile’s corrector formula is
𝒉
𝒚𝟒, 𝒄 = 𝒚𝟐 + [𝒚 + 𝟒𝒚𝟑 + 𝒚𝟒 ]
𝟑 𝟐
0.1
𝑦 , = 1.0097 + [0.0467 + 4(0.0452 + 0.0437]
3

𝑦 , = 1.01874

4(0.1)
=1+ [2(1.3552) − 1.8535 + 2(2.6589)]
3
𝑦 , = 1.8233
𝑦 =𝑥 𝑦 +𝑦 = (0.4)(1.8233) + (1.8233) = 4.0537
By Mile’s corrector formula is
ℎ
𝑦 , = 𝑦 + [𝑦 + 4𝑦 + 𝑦 ]
3
0.1
𝑦 , = 1.2774 + [1.8535 + 4(2.6589) + 4.0537]
3
𝑦 , = 1.8165
𝒅𝒚
10. Using Runge-kutta method of fourth order, find y for 𝒙 = 𝟎. 𝟏, 𝟎. 𝟐, 𝟎. 𝟑 given that = 𝒙𝒚 +
𝒅𝒙

𝒚𝟐 , 𝒚(𝟎) = 𝟏 Continue the solution at x=0.4 using Milne’s method .

Solution:

Given = 𝑥𝑦 + 𝑦 , 𝑥 = 0, 𝑦 = 1, ℎ = 0.1
By Runge –kutta method
𝑘 = ℎ𝑓(𝑥 , 𝑦 )
= (0.1)𝑓(0,1)
= (0.1)(𝟎 + 𝟏)
𝑘 = 0.1
ℎ 𝑘
𝑘 = ℎ𝑓 𝑥 + , 𝑦 +
2 2
0.1 0.1
= (0.1)𝑓 0 + ,1 +
2 2
= (0.1)𝑓(0.05,1.05)
= (0.1) (0.05)(1.05) + (1.05)
𝑘 = 0.1155
ℎ 𝑘
𝑘 = ℎ𝑓 𝑥 + , 𝑦 +
2 2
0.1 0.1155
= (0.1)𝑓 0 + ,1 +
2 2
= (0.1)𝑓(0.05,1.50775)
= (0.1) (0.05)(1.50775) + (1.50775)
𝑘 = 0.1172
𝑘 = ℎ𝑓(𝑥 + ℎ, 𝑦 + 𝑘 )
= (0.1)𝑓(0 + 0.1,1 + 0.1172)
103
 = (0.1)𝑓(0.1,1.4424)
= (0.1) (0.1)(1.4424) + (1.4424)
𝑘 = 0.1260
1
∆𝑦 = (𝑘 + 2𝑘 + 2𝑘 + 𝑘 )
6
1
= (0.1 + 2(0.1155) + 2(0.1172) + 0.1260)
6
∆𝑦 = 0.1152
𝑦 = 𝑦 + ∆𝑦
= 1 + 0.1152
𝒚(𝟎. 𝟏) = 𝟏. 𝟏𝟏𝟓𝟐
To find y(0.2):
Here 𝑥 = 0.1, 𝑦 = 1.1152

𝑘 = ℎ𝑓(𝑥 , 𝑦 )
= (0.1)𝑓(0.1,1.1152)
= (0.1) (0.1)(1.1152) + (1.1152)
𝑘 = 0.1255
ℎ 𝑘
𝑘 = ℎ𝑓 𝑥 + , 𝑦 +
2 2
0.1 0.1255
= (0.1)𝑓 0.1 + , 1.1152 +
2 2
= (0.1)𝑓(0.05,1.1780)
= (0.1) (0.05)(1.1780) + (1.1780)
𝑘 = 0.1355
ℎ 𝑘
𝑘 = ℎ𝑓 𝑥 + , 𝑦 +
2 2
0.1 0.1355
= (0.1)𝑓 0.1 + ,1 +
2 2
= (0.2)𝑓(0.05,1.1355)
= (0.1) (0.05)(1.1355) + (1.1355)
𝑘 = 0.1577
𝑘 = ℎ𝑓(𝑥 + ℎ, 𝑦 + 𝑘 )
= (0.1)𝑓(0.1 + 0.1,1.1152 + 0.1577)
= (0.1)𝑓(0.2,1.2729)
= (0.1) (0.2)(1.2729) + (1.2729)
𝑘 = 0.1875
1
∆𝑦 = (𝑘 + 2𝑘 + 2𝑘 + 𝑘 )
6
1
= (0.1255 + 2(0.1355) + 2(0.1577) + 0.1875)
6
∆𝑦 = 0.1499
𝑦 = 𝑦 + ∆𝑦
= 11152 + 0.1499
𝒚(𝟎. 𝟐) = 𝟏. 𝟐𝟔𝟓𝟏

104
 To find y(0.3):
Here 𝑥 = 0.2, 𝑦 = 1.2651
𝑘 = ℎ𝑓(𝑥 , 𝑦 )
= (0.1)𝑓(0.2,1.2651)
= (0.1) (0.2)(1.2651) + (1.2651)
𝑘 = 0.1853
ℎ 𝑘
𝑘 = ℎ𝑓 𝑥 + , 𝑦 +
2 2
0.1 0.1
= (0.1)𝑓 0.2 + , 1.2651 +
2 2
= (0.1)𝑓(0.25,1.3578)
= (0.1) (0.25)(1.3578) + (1.3578)
𝑘 = 0.2183
ℎ 𝑘
𝑘 = ℎ𝑓 𝑥 + , 𝑦 +
2 2
0.1 0.2183
= (0.1)𝑓 0.2 + , 1.2651 +
2 2
= (0.1)𝑓(0.25,1.3742)
= (0.1) (0.25)(1.3742) + (1.3742)
𝑘 = 0.2232
𝑘 = ℎ𝑓(𝑥 + ℎ, 𝑦 + 𝑘 )
= (0.1)𝑓(0.2 + 0.1,1.2651 + 0.2232)
= (0.1)𝑓(0.3,1.4883)
= (0.1) (0.1)(1.4883) + (1.4883)
𝑘 = 0.2662
1
∆𝑦 = (𝑘 + 2𝑘 + 2𝑘 + 𝑘 )
6
1
= (0.1853 + 2(0.2183) + 2(0.2232) + 0.2662)
6
∆𝑦 = 0.2224
𝑦 = 𝑦 + ∆𝑦
= 1.2651 + 0.2224
𝑦(0.3) = 1.4875
𝑥 = 0, 𝑥 = 0.1, 𝑥 = 0.2, 𝑥 = 0.3, 𝑥 = 0.4
𝑦 = 1, 𝑦 = 1.1152, 𝑦 = 1.2651, 𝑦 = 1.4875, 𝑦 =?
𝑦′ = 𝑥𝑦 + 𝑦
𝑦 = 𝑥 𝑦 + 𝑦 𝟐 = (0)(1) + (1)𝟐 = 𝟏
𝑦 = 𝑥 𝑦 + 𝑦 𝟐 = (0.1)(1.1152) + (1.1152)𝟐 = 1.3552
𝑦 = 𝑥 𝑦 + 𝑦 𝟐 = (0.2)(1.2651) + (1.2651)𝟐 = 1.8535
𝑦 = 𝑥 𝑦 + 𝑦 𝟐 = (0.3)(1.4875) + (1.4875)𝟐 = 2.6589
By Mile’s predictor formula is
𝟒𝒉
𝒚𝟒, 𝒑 = 𝒚𝟎 + [𝟐𝒚𝟏 − 𝒚𝟐 + 𝟐𝒚𝟑 ]
𝟑
4(0.1)
𝑦 , =1+ [2(1.3552) − 1.8535 + 2(2.6589)]
3
𝑦 , = 1.8233
105
 𝑦 = 𝑥 𝑦 + 𝑦 𝟐 = (0.4)(1.8233) + (1.8233)𝟐 = 4.0537
By Mile’s corrector formula is
𝒉
𝒚𝟒, 𝒄 = 𝒚𝟐 + [𝒚𝟐 + 𝟒𝒚𝟑 + 𝒚𝟒 ]
𝟑
0.1
𝑦 , = 1.2651 + [1.8535 + 4(2.6589) + 4.0537]
3

𝑦 , = 1.8165

106