STATISTICS AND NUMERICAL METHODS

QUESTION II NOVEMBER / DECEMBER 2010

1. What are the parameters and statistics in sampling?
Solution:
To avoid verbal confusion with the statistical constant of the population, namely which are usually
referred to as parameters. Statistical measures computed from sample observations alone. E.g.
are usually referred to as statistic.

2. Write any two applications of test.

Solution:
is used to test whether differences between observed and expected frequencies or significances.
3. Compare One way classification modal with Two way classification modal.
Solution:
One way Two way
We cannot test two sets Two sets of hypothesis
1
of Hypothesis can tested.
Data are classified according Data are classified according
2
to one factor to two different factor.

4. What is meant by Latin square?

Solution:
The treatments are then allocated at random to these rows and columns in such a way that every treatment occurs once
and only once in each row and in each column. Such a layout is known as Latin square design.
5. Write the convergence condition and order of convergence for Newton- Raphson –method.
Solution:
The Criterion for convergence of Newton- Raphson -method is

The order of convergence of Newton- Raphson –method is 2.

6. Compare Gauss Jacobi with Gauss Jordan.
Solution:

Gauss – Jordan Gauss – Jacobi method

1 It gives exact value Convergence rate is slow
2 Simple, it takes less time Indirect method
This method determines all the roots at This method determines only one root at
3
the same time. a time.

7. Create a forward difference table for the following data and state the degree of polynomial for the same.
Solution:
8. Compare Simpson’s rule with Trapezoidal method.

Solution:

Simpson’s Rule Trapezoidal rule

1 Must accurate Least accurate
Interval of integration must be divided Can be divide into any number of
2
into even number of subintervals subintervals.
9. Using Taylor series find

Solution:

Taylor series formula is

Therefore equation (1) becomes,

10. Solve .
Solution:
Given

.
.
Part - B
11. (a). (i). A machine puts out 16 imperfect articles in a sample of 500. After the machine is overhauled, it puts out 3 imperfect
articles in a batch of 100. Has the machine been improved?
Solution:
Null Hypothesis:

Alternative hypothesis:

Here Population Proportion P is not known.

At 5% level of significance, the table value for .

Conclusion: We accept the null hypothesis. That is the machine has improved after service.

11 (a). (ii). Examine whether the difference in the variability in yields is significant at 5% level of significance, for the following.

Set of 40 plots Set of 60 plots

Mean yield per plot 1256 1243
S.D per plot 34 28
Solution:
Given that

Null Hypothesis: i.e., there is no difference b/w two sets of yields.

Null Hypothesis:

The test statistic is given by

At 5% significance level the tabulated value for is 1.96.

Conclusion:

, we reject the Null Hypothesis. That is the two plots of yields differs significantly.

11. (b). (i). Test if the difference in the means is significantly for

Sample I 76 68 70 43 94 68 33
Sample II 40 48 92 85 70 76 68 22
Solution:

Hence

Null Hypothesis :
i.e., there is no significant difference between two samples. groups.
Alternative Hypothesis :
The test statistic is given by

The table value for at 5% level of significance with degrees of freedom is .

Calculated value =0.242 and Tabulated value =2.16

Conclusion:
 , we Accept . That is there is no significant difference between the two sample means.

11. (b). (ii) The following data gives the number of aircraft accidents that occurred during the various days of a week. Find
whether the accidents are uniformly distributed over the week.
Days Sun Mon Tue Wed Thu Fri Sat
No. of Accidents 14 16 8 12 11 9 14
Solution:
Null Hypothesis The accidents are uniformly distributed over the week.
Alternative Null Hypothesis The accidents are not uniformly distributed over the week.
The test statistic is given by

Observed Expected
Day
freq freq
Sun 14 12 2 4 0.333333
Mon 16 12 4 16 1.333333
Tue 8 12 -4 16 1.333333
Wed 12 12 0 0 0
Thu 11 12 -1 1 0.083333
Fri 9 12 -3 9 0.75
Sat 14 12 2 4 0.333333
84 84 4.166

Table value of with d.o.f is 12.59.

Conclusion:
Since , we accept null hypothesis. That is the air accidents are uniformly distributed over the week.

12 (a). Carry out the 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: Let us take the null hypothesis that

1. The 5 workers do not differ with respect to mean productivity

2. The mean productivity is the same for the four different machines.
To simplify calculation let us subtract 40 from each value, the new values are

Machine Type
A B C D Total
1 4 -2 7 -4 5
2 6 0 12 3 21
Workers 3 -6 -4 4 -8 -14
4 3 -2 6 -7 0
5 -2 2 9 -1 8
Total 5 -6 38 -17 20

Between column sum of squares

Between Row sum of squares

Error sum of squares

ANOVA table for two-way classification

Source of Degrees of Sum of squares Table Value

Mean sum of squares (MS) F-Ratio
variation freedom (SS)

B/w
Column

B/w Row

Error

Conclusion:

1. . Hence is accepted. That is the 5 workers differ respect to mean productivity.

 2. . Hence is rejected. That is the mean productivity is not the same for the four machines.

12. (b) Perform Latin square experiment for the following:

Roam I II III Three equally spaced concentrations of
poison as extracted from the scorpion
Arabic 1 2 3 Three equally spaced body weights for the animals tested.
Latin A B C Three equally spaced times of storage of the poison before it
is administered to the animals.
I II III
1 0.194 0.73 1.187
A B C
2 0.758 0.311 0.589
C A B
3 0.369 0.558 0.311
B C A
Solution:
Null hypothesis: There is no significant difference between rows, columns and between the treatments.

1 2 3 Total
1 0.194 0.73 1.187 1.485
2 0.758 0.311 0.589 0.916
3 0.369 0.558 0.311 0.511
Total 1.321 1.599 2.087 5.007

Between Column sum of squares

Between Row sum of squares

Treatment sum of squares SSK

Treatment total

Error sum of squares

Degrees of freedom:

ANOVA table for three-way classification Table

Value
Source of Sum of squares
D.o.f Mean sum of squares (MS) F-Ratio
variation (SS)

B/W
2 SSC=0.096
Column

B/W Row 2 SSR=0.122

B/w
2 SSK=0.47
Treatment

Error 2 SSE=0.064

Conclusion:

Accept . That is there is no difference between Row, Column and Treatments.

13. (a). (i). Find the inverse of

Solution:
While we find the inverse of the matrix A, the diagonal elements should not be zero. If it is zero, then we rearrange the
given matrix in correct form.

We know that

Now,
Now, we need to make as a diagonal matrix.

13. (a). (ii). Solve by Gauss Siedal method

6
Solution: 6

Since the diagonal elements are not dominant in the coefficient matrix, we rewrite the given equation as follows as follows

6
From the above equation, we have

Gauss Siedal Method:

We form the Iterations in the table
Iteration
 2.5 2.09 1.14
2.99 2.05 0.909
3.04 1.98 0.90
3.02 1.98 0.91
3.02 1.98 0.91
3.02 1.98 0.91
Hence the solution is

13. (b). (i) Using Gauss Jordan method, solve the following system
.
Solution: Let the given system of equations be

The given system is equivalent to

Now, we need to make as a diagonal matrix.

Which is a diagonal matrix, from the matrix, we have

13. (b) (ii). Find all the Eigen value of using power method.

Using as initial vector.

 Solution : Let be the initial vector.

Therefore,

The dominant Eigen value = 4.

Corresponding Eigen vector is .

14. (a). (i). Taking , evaluate by Simpson’s rule. Verify the answer with integration.

Solution:
Here . We form a table

Simpson’s 1/3 rule :

By Actual Integration:
.

14. (a). (ii). Use Lagrange’s formula to fit a polynomial to the following data hence find .

Solution:

Lagrange’s interpolation formula, we have

14. (b) (ii). Evaluate using Trapezoidal rule. Verify the answer with direct integration.

Solution:

Here . Range

So we divide 6 equal intervals with .

We form a table
 .038462

Trapezoidal rule

By Actual Integration:

14. (b). (ii) Find from the following

X: 1941 1951 1961 1971 1981 1991

Y: 20 24 29 36 46 51
Solution: We form the difference table

1941 20
4
1951 24 1
5 1
1961 29 2 0
7 1
1971 36 3 1
10 2
1981 46 5
5
1991 51
The Newton’s backward formula is

The Newton’s backward interpolation formula is

Here
15. (a) (i). Use Modified Euler’s method, with to find the solution of with in .

Solution:

Given , .

The Modified Euler’s formula is

To find

Put equation becomes

15. (a). (ii). Using Milne’s method, obtain the solution of given

Solution: Given
 and

The Milne’s Predictor formula is

To Find y(0.8) :

Put n=3 in equation (1), we have

Equation (2) becomes

The Milne’s Corrector formula is

Put n=3 in equation (3), we have

Equation (4) becomes

To Find :

Put n=3 in equation (1), we have

Equation (2) becomes

The Milne’s Corrector formula is

Put n=3 in equation (3), we have

Equation (4) becomes

Result:

15. (b). (i). Use R.K method fourth order to the if .

Solution:

The algorithm for fourth order R-K method is

To find

To find

15. (b). (ii) Solve .

Solution:

The given difference equation is

The auxiliary equation is