Session: 2018-2019

1. A. Resolve the following system using the Cramer’s Rule:

0.40x1-0.1x2-0.2x3=7.85
0.10x1+7x2-0.3x3=-19.3
0.30x1-0.2x2+10x3=71.4
Soln: Let us write these equations in the form AX=B
. 4 −.1 −.2 𝑥1 7.85
[. 1 7 −.3] [𝑥2 ] = [−19.3]
. 3 −.2 10 𝑥3 71.4
.4 −.1 −.2
D=|A|=|. 1 7 −.3|=28.509
.3 −.2 10
7.85 −.1 −.2
Dx1=|−19.3 7 −.3|=631.059
71.4 −.2 10
. 4 7.85 −.2
Dx2=|. 1 −19.3 −.3|=-79.7745
. 3 71.4 10
. 4 −.1 7.85
𝐷𝑥3 = |. 1 7 −19.3|=183.027
. 3 −.2 71.4
X1= Dx1/D=631.059/28.509≈22.135
X2= Dx2/D=-79.7745/28.509≈-2.79
X3= 𝐷𝑥3 /D=183.027/28.509≈6.41
B. use the Gauss-Elimination technique to resolve the following system:
0.3x1-0.1x2-0.2x3=7.85
0.10x1+7x2-0.3x3=-19.3
x1-0.2x2+10x3=71.4
Soln: We'll set up the augmented matrix [A|B]:
A|B = | 0.3 -0.1 -0.2 | 7.85 |
| 0.10 7 -0.3 | -19.3 |
| 1 -0.2 10 | 71.4 |
Now, let's perform row operations to solve the system:
Step 1: Make the coefficient of the first row's first element (A[1,1]) equal to 1. Divide the
first row by 0.3:
A|B = | 1 -0.3333 -0.6667 | 26.1667|
| 0.10 7 -0.3 | -19.3 |
| 1 -0.2 10 | 71.4 |

Step 2: Eliminate the coefficients below A[1,1]. Multiply the first row by -0.10 and add it
to the second row. Multiply the first row by -1 and add it to the third row:
A|B = | 1 -0.3333 -0.6667 | 26.1667 |
| 0 7.03333 -0.2333 | -21.4833 |
| 0 0.1333 10.6667 | 45.2333 |

Step 3: Make the coefficient of the second row's second element (A[2,2]) equal to 1.
Divide the second row by 7.3333:
A|B = | 1 -0.3333 -0.6667 | 26.1667 |
| 0 1 -0.0318 | -2.9256 |
| 0 0.1333 10.6667 | 45.2333 |

Step 4: Eliminate the coefficients below A[2,2]. Multiply the second row by -0.1333 and
add it to the third row:
A|B = | 1 -0.3333 -0.6667 | 26.1667 |
| 0 1 -0.0318 | -2.9256 |
| 0 0 1 | 3.7866 |
Now, the augmented matrix is in row-echelon form, and we can back-substitute to find
the solutions:
From the third row, we have:
1x3 = 3.7866
x3 = 3.7866

From the second row, we have:

1x2 -0.0318x3= -2.9256
x2 =-2.80

From the first row, we have:

1x1 - 0.3333x2 -0.6667x3 = 26.1667
x1 = 27.71
 So, the correct solution to the system of equations is:
x1 ≈ 28
x2 ≈ -2.80
x3 ≈ 3.7866
2. A. Apply the Cholesky’s process to locate the root of the following system
X1-X2-X3=2
2X1+3X2+5X3=-3
3X1+2X2-3X3=6.
n
o Sol : Rules or technique→ AX=B => LUX=B => LY=B then UX=Y
1 −1 −1 𝑥1 2
A=[2 3 5 ] , 𝑋 = [𝑥2 ] , 𝐵 = [−3]
3 2 −3 𝑥3 6
A matrix can be written as the product of a lower triangular matrix and an upper
triangular matrix
1 −1 −1 𝑙11 0 0 𝑢11 𝑢12 𝑢13
A=[2 3 5 ] = [𝑙21 𝑙22 0 ] [ 0 𝑢22 𝑢23 ]
3 2 −3 𝑙31 𝑙32 𝑙33 0 0 𝑢33
1 0 0 𝑢11 𝑢12 𝑢13
= [𝑙21 1 0] [ 0 𝑢22 𝑢23 ]
𝑙31 𝑙32 1 0 0 𝑢33
𝑢11 𝑢12 𝑢13
=[𝑙21 𝑢11 𝑙21 𝑢12 + 𝑢22 𝑙21 𝑢13 + 𝑢23 ]
𝑙31 𝑢11 𝑙31 𝑢12 + 𝑙32 𝑢22 𝑙31 𝑢13 + 𝑙32 𝑢23 + 𝑢33
Comparing the values,
u11=1, u12=-1, u13=-1, l21u11=2, so that l21=2
Similarly, we can find out another unknown variable,
AX=B => LUX=B
1 0 0 1 1 −1 2
[2 1 0] [0 1 7 ] X=[−3]
3 −1 1 0 0 7 6
Let, UX=Y => LY=B
1 0 0 𝑦1 2
𝑦
[2 1 0] [ 2 ]= [−3]
3 −1 1 𝑦3 6
From above matrix we find,
✓ y1=2
✓ 2y1+ y2=-3
 => y2=-7
✓ 3y1- y2+y3=6=> y3=-7
Again, UX=Y
1 1 −1 𝑥1 2
𝑥
[0 1 7 ] [ 2 ]= [−7]
0 0 7 𝑥3 −7
From above matrix we find,
✓ 7x3=-7
x3=-1
✓ x2+ 7x3=-7
=> x2=0
✓ x1+ x2-Y3=2
=> x1=1
B.
i. How can we eliminate the error in the Trapezoidal rule by applying the Simpson’s
rule?
To eliminate error in the Trapezoidal rule using Simpson's rule, follow these steps:
▪ Divide the Interval: Divide the given interval into an even number of
subintervals (n) for Simpson's rule. The Trapezoidal rule may use any number
of subintervals, but Simpson's rule requires an even number.
▪ Apply Trapezoidal Rule: Use the Trapezoidal rule to approximate the integral
over each subinterval individually. This gives you an estimate of the integral
using linear approximations.
▪ Apply Simpson's Rule: Now, apply Simpson's rule over the same subintervals.
Simpson's rule uses quadratic approximations, which can provide a more
accurate estimate.
▪ Calculate the Difference: Subtract the result from Simpson's rule from the
result from the Trapezoidal rule for each subinterval.
▪ Sum the Differences: Sum up all the differences calculated in step 4 for each
subinterval.
 ii. Integrate using Simpson’s 3/8 rule.
f(x)=0.2+25x-200x2+675x3-900x4+400x5 from a=0 to b=0.8
soln: A single application of Simpson’s 3/8 rule requires four equally spaced points:
➢ Points and function values:
f(0)=0.2
𝑎+𝑏 0+0.8
= =0.2667
3 3
f(0.2667)=1.432724
0.2667+0.2667=0.5333
f(0.5333)=3.487177
0.5333+0.2667=0.8
f(0.8)=0.232
➢ Calculation of approximate integral:
using Simpson’s 3/8 rule we find,
0.2+3(1.432724+3.487177)+0.232
I=0.8 X =1.519170
8
➢ Error Calculation:
The true integral value is 1.640533.
So, Et = |True Value - Approximate Value|
= |1.640533 - 1.519170|
= 0.121363.
(Et / True Value) x 100% = (0.121363 / 1.640533) x 100% ≈ 7.4%.
3. A. Solve the following system using the Gauss-Jordan Method:
0.7X1-0.1X2-0.2X3=7.85
0.1X1+7X2-0.3X3=-19.3
X1-0.2X2+10X3=71.4
Starting with the augmented matrix:
[ 0.7 -0.1 -0.2 | 7.85 ]
[ 0.1 7 -0.3 | -19.3 ]
[ 1 -0.2 10 | 71.4 ]

Step 1: Divide Row 1 by 0.7:

[ 1 -0.1429 -0.2857 | 11.2143 ]
[ 0.1 7 -0.3 | -19.3 ]
[1 -0.2 10 | 71.4 ]
Step 2: Subtract 0.1 times Row 1 from Row 2:
[ 1 -0.1429 -0.2857 | 11.2143 ]
[ 0 7.1429 -0.2857 | -20.1143 ]
[ 1 -0.2 10 | 71.4 ]

Step 3: Subtract Row 1 from Row 3:

[ 1 -0.1429 -0.2857 | 11.2143 ]
[ 0 7.1429 -0.2857 | -20.1143 ]
[ 0 0.0571 10.2857 | 60.1857 ]

Step 4: Divide Row 2 by 7.1429:

[ 1 -0.1429 -0.2857 | 11.2143 ]
[0 1 -0.0399 | -2.8157 ]
[ 0 0.0571 10.2857 | 60.1857 ]

Step 5: Subtract 0.0571 times Row 2 from Row 3:

[ 1 -0.1429 -0.2857 | 11.2143 ]
[0 1 -0.0399 | -2.8157 ]
[0 0 10.2857 | 61.7571 ]

Step 6: Divide Row 3 by 10.2857:

[ 1 -0.1429 -0.2857 | 11.2143 ]
[0 1 -0.0399 | -2.8157 ]
[0 0 1 | 5.9999 ]

Step 7: Add 0.0399 times Row 3 to Row 2:

[ 1 -0.1429 -0.2857 | 11.2143 ]
[0 1 0 | -2.68 ]
[0 0 1 | 5.9999 ]

Step 8: Add 0.2857 times Row 3 to Row 1:

[ 1 -0.1429 0 | 12.498 ]
[0 1 0 | -2.68 ]
[0 0 1 | 5.9999 ]
Step 9: Add 0.1429 times Row 1 to Row 2:
[1 0 0 | 12.498 ]
[0 1 0 | -2.68 ]
[0 0 1 | 5.9999 ]

So, the solution to the system of equations is:

X1 = 12.498
X2 = -2.68
X3 = 5.9999

Rounded to a reasonable number of decimal places, the solution is:

X1 ≈ 12.5
X2 ≈ -2.68
X3 ≈ 6.0
B.
i. Show the scenarios. In the case of an iteration process, there is convergence
and divergence.
▪ Convergence:
• Numerical Approximation: Imagine you are using the Newton-Raphson
method to find the root of a mathematical function.
• Gradient Descent (Optimization): In machine learning, gradient descent is
often used to minimize a loss function.
• Power Iteration (Eigenvalues): When finding the largest eigenvalue of a
matrix using the power iteration method, the sequence of approximations
will often converge to the largest eigenvalue and its corresponding
eigenvector.
▪ Divergence:
• Numerical Instability: In some cases, numerical instability can lead to
divergence. For instance, when solving a system of linear equations using
an iterative method, if the matrix is ill-conditioned, the solution may start
to diverge instead of converging to the correct answer.
• Learning Rate Too High (Gradient Descent): In machine learning, if the
learning rate in gradient descent is set too high, it can cause the
optimization process to diverge.
• Chaos Theory: In some complex systems described by chaos theory, even
small changes in initial conditions can lead to divergent behavior
 ii. Write down the algorithm of Bisection Method.
1. Choose 2 real numbers a & b such that f(a)*f(b)<0.
𝑎+𝑏
2. Define root, c= .
2
3. Find f(c).
4. If f(a)xf(c)<0
then set b=c
else set a=c
return the step-1 until finding the root matched twice.
4 [A] Fit a second-order polynomial to the data of the following table. Also find out
standard error Sy/x.
Xi Yi
0 2.1
1 7.7
2 13.6
∑ 𝒙𝒊 = 𝟑 ∑ 𝒚𝒊 = 𝟐𝟑. 𝟒

m=2, n=3, ∑ 𝑥𝑖 = 3 , ∑ 𝑥𝑖4 = 17

∑ 𝑥𝑖2 = 5 , ∑ 𝑥𝑖3 = 9, ∑ 𝑥𝑖𝑦𝑖 =34.9
∑ 𝑥𝑖2 𝑦𝑖 = 62.1, ∑ 𝑦𝑖 = 23.4
∑ 𝑥𝑖
𝑥̅ = =3/3=1
𝑛
∑ 𝑦𝑖
𝑦̅= =23.4/3=7.8
𝑛
Therefore, the simultaneous Linear equations are:
3 3 5 𝑎0 23.4
[3 5 9 ] [𝑎1]=[34.9]
5 9 17 𝑎2 62.1
Solving these equations through a technique such as Gauss-elimination gives a0=7.55
a1=-2.725 and a2=2.875

=0.14332+1.00286+1.08158=2.22776
 2.22776
=√
3−(2+1)

2.22776
=√
0
= Something/divided by zero.
=Undefined.
Unfortunately, the results are undefined.

4 B. Derive equation for linear regression and find a0 and a1.

The simplest example of a least-squares approximation is fitting a straight line to a set of
paired observations: (x1, y1), (x2, y2), . . . , (xn, yn).
The mathematical expression for the straight line is:
y = a0 + a1x + e---------------------------------------------------------17.1
where a0 and a1 are coefficients representing the intercept and the slope, respectively,
and e is the error, or residual, between the model and the observations, which can be
represented by re-arranging Eq. (17.1) as
e = y − a0 − a1x
➢ Criteria for a “best” fit: One strategy for fitting a “best” line through the data would
be to minimize the sum of the residual errors for all the available data, as in

✓ where n = total number of points

Therefore, another logical criterion might be to minimize the sum of the absolute
values of the discrepancies, as in

A strategy that overcomes the shortcomings of the aforementioned approaches is to

minimize the sum of the squares of the residuals between the measured y and the y
calculated with the linear model
➢ least-square fit of a straight line: To determine values for a0 and a1, Eq. (17.3) is
differentiated with respect to each coefficient:

Note that we have simplified the summation symbols; unless otherwise indicated, all
summations are from i = 1 to n. Setting these derivatives equal to zero will result in a
minimum Sr. If this is done, the equations can be expressed as

Now, realizing that ∑ a0 = na0, we can express the equations as a set of two
simultaneous linear equations with two unknowns (a0 and a1):