Maths NA 2

2
Unit IN T E R PO LA T iO N A ND
P O LYNO M iA L APPR O X iM A T iON
LEARNiNG OBJEC TiVES

After studying this unit, a student will have thorough knowledge about the following key concepts.
v Interpolation and Lagrange Polynomial
v Data Approximation and Neville's Method
v Divided differences
v Hermite Interpolation
v Cubic Spline Interpolation.
INT RODUC TiON

Interpolation is the method of estimating new data points between well known discrete points .It is
the process of making most appropriate estimate for the missing data. In engineering and science, the
number of data points obtained by sampling or experimentation represents the value of a function
for a limited number of values of the independent variable. This value of the function is estimated or
interpolated for an intermediate value of the independent variable. Many mathematicians contributed
in the field of interpolation like Gauss, Newton, Bessel, Stirling etc. Interpolation technique is used in
various disciplines like Engineering, Business, Economics etc.
Mathematics Paper-VII Numerical Analysis
Part-A
Short Questions with Solutions
Q1. Let f(x) = ex (0 ≤ x ≤ 2), approximate f(0.25) using linear interpolation with x0 = 0 and x1 = 0.5.
Answer : Model Paper-1, Q3
Given function is,

f(x) = ex in [0, 2]
And
x0 = 0, x1 = 0.5
The linear interpolation is given as,
P1(x) = L0(x) f(x0) + L1(x)f(x1) ... (1)
Where,
x–x
L0(x) = x – x1
0 1
x – 0.5
= 0 – 0.5
⇒ L0(x) = x––00.5.5
x–x
L1(x) = x – x0
1 0
x–0
= 0.5 – 0
⇒ L1(x) = 0x.5
f(x0) = f(0) = e0 = 1
f(x1) = f(0.5) = e0.5 = 1.6487212

Substituting the corresponding values in equation (1),
J N
P1(x) = KK x––00.5.5 OO (1) + ` 0x.5 j (1.6487212)
L P
= –2(x – 0.5) + 3.297442x
= –2x + 1 + 3.297442x
⇒ P1(x) = 1.2974424x + 1
P1(0.25) = 1.297442(0.25) + 1
⇒ P1(0.25) = 1.32436
P1(0.25) is an approximation of f(0.25).
\ f(0.25) = e0.25 = 1.28402.

Q2. Let f(x) = ex(0 ≤ x ≤ 2), approximate f(0.75) using linear interpolation with x0 = 0.5 and x1 = 1.
Given function is,

f(x) = ex in [0, 2]
Linear interpolation is given as,
P1(x) = L0(x) f(x0) + L1(x) f(x1) ... (1)
88 SIA PUBLISHERS and DISTRIBUTORS PVT. LTD.
UNIT-2: Interpolation and Polynomial Approximation Mathematics Paper-VII
Where,
x–x
L0(x) = x – x1
0 1
= 0x.5––11
⇒ L0(x) = –x 0–.15
x–x
L1(x) = x – x0
1 0
= 1x –– 00..55
⇒ L1(x) = x 0–.50.5
f(x0) = f(0.5) = e0.5 = 1.64872
f(x1) = f(1) = e1 = 2.71828.

J N J N
P1(x) = KK –x 0–.15 OO (1.64872) + KK x 0–.50.5 OO (2.71828)
L P L P
= –3.29744(x – 1) + 5.43656(x – 0.5)
= –3.29744 x + 3.29744 + 5.43656 x – 2.71828
⇒ P1(x) = 2.13912 x + 0.57916

P1(0.75) = 2.13912(0.75) + 0.57916
⇒ P1(0.75) = 2.1835
P1(0.75) is an approximation of f(0.75).
\ f(0.75) = e0.75 = 2.117.

Q3. Let f(x) = ex(0 ≤ x ≤ 2), approximate f(0.25) and f(0.75) by using the second interpolating polynomial with
x0 = 0, x1 = 1, x2 = 2.
Given function is,

f(x) = ex in [0, 2]
f(x0) = f(0) = e0 = 1
f(x1) = f(1) = e1 = 2.718281.
f(x2) = f(2) = e2 = 7.38905.

The second interpolating polynomial is given as,
P2(x) = L0(x) f(x0) + L1(x)f(x1) + L2(x)f(x2) ... (1)
Where,
(x – x ) (x – x )
L0(x) = (x – x1) (x – x2 )
0 1 0 2
(x – 1) (x – 2)
= (0 – 1) (0 – 2)
( x – 1 ) ( x – 2)
⇒ L0(x) = 2
SIA PUBLISHERS and DISTRIBUTORS PVT. LTD. 89
(x – x ) (x – x )
L1(x) = (x – x0) (x – x2 )
1 0 1 2
(x – 0) (x – 2)
= (1 – 0) (1 – 2)
x (x – 2)
⇒ L1(x) = –1
(x – x ) (x – x )
L2(x) = (x – x0) (x – x1 )
2 0 2 1
( x – 0) ( x – 1 )
= (2 – 0) (2 – 1)
x ( x – 1)
⇒ L2(x) = 2
(x – 1) (x – 2) x (x – 2) x ( x – 1)
P2(x) = 2 (1) + –1 (2.718281) + 2 (7.38905)
= (x2 – 3x + 2) (0.5) – 2.71828 (x2 – 2x) + 3.69452 (x2 – x)
= x2 (0.5 – 2.71828 + 3.69452) + x(– 1.5 + 5.43656 – 3.69452) + 1

⇒ P2(x) = x2 (1.47624) + x(0.24204) + 1
P2(0.25) = (1.47624) (0.25)2 + (0.24204) (0.25) + 1
= 1.152775
P2(0.75) = (1.47624) (0.75)2 + (0.24204) (0.75) + 1
= 2.011915
\ f(0.25) = e0.25 = 1.28402
f(0.75) = e0.75 = 2.117.

Q4. Using Neville’s method, determine Q1,1 at x = 1.5, if f(1) = 0.6931 and f(1.3) = 0.7885.
Given,
f(1) = 0.6931, f(1.3) = 0.7885
From Neville’s method, Q1,1 is given as,
(x – x0) Q1,0 – (x – x1) Q0,0
Q1,1(x) = x1 – x0
Here,
x0 = 1, x1 = 1.3
Q0,0 = 0.6931 and Q1,0 = 0.7885.
(1.5 – 1) (0.7885) – (1.5 – 1.3) (0.6931)
⇒ Q1,1 (1.5) = 1.3 – 1
(0.5) (0.7885) – (0.2) (0.6931)
= 0.3
= 0.8521
\ Q1,1 (1.5) = 0.8521

Q5. Find the value of p0,1,2,3(1.5) if p0,1 (x) = x+ 1, p1,2(x) = 3x – 1 and p1,2,3(1.5) = 4 for xj = j where j = 0,1,2,3.
Given that,
P0, 1(x) = x + 1
P1, 2(x) = 3x – 1
P1, 2, 3(1.5) = 4
For xj = j, j = 0,1 , 2, 3
x0 = 0, x1 = 1, x2 = 2, x3 = 3.
P0, 1(1.5) = 1.5 + 1 ⇒ P0, 1(1.5) = 2.5
P1, 2(1.5) = 3(1.5) – 1 ⇒ P1, 2(1.5) = 3.5
(x – x0) P1, 2 (x) – (x – x2) P0, 1 (x)
P0, 1, 2(x) = x2 – x0
(1.5 – 0) 3.5 – (1.5 – 2) (2.5)

⇒ P0, 1, 2(1.5) = 2–0
⇒ P0, 1, 2(1.5) = 3.25
(x – x0) P1,2,3 (x) – (x – x3) P0, 1,2 (x)
P0, 1, 2,3 (1.5) = x –x
3 0
(1.5 – 0) P1,2,3 (1.5) – (1.5 – 3) P0, 1,2 (1.5)

P0, 1, 2,3 (1.5) = 3–0
(1.5 – 0) (4) – (1.5 – 3) (3.25)
= 3
= 3.625
\ P0, 1, 2, 3(1.5) = 3.625
Q6. Using Newton’s divided difference formula, construct the interpolating polynomial of degree three for
the unequal spaced points given in the following table.
x f(x)
– 0.1 5.30000
0.0 2.00000
0.2 3.19000
0.3 1.00000
Given,
x0 = – 0.1, x1 = 0.0, x2 = 0.2, x3 = 0.3
f(x0) = 5.30000, f(x1) = 2.00000, f(x2) = 3.19000, f(x3) = 1.00000
The first divided difference corresponding to x0 and x1 is given as,

f [x1] – f [x0]
f [x0, x1] = x1 – x0
= 2.00000 – 5.30000
0.0 + 0.1
f [x0, x1] = – 33
The remaining first divided differences can be calculated as shown in the table.
x f(x) First divided difference Second divided difference Third divided difference
– 0.1 5,30000
f(x0, x1)
– 33 f(x0, x1, x2)
0.0 2.00000 129.8333
f(x1, x2) f(x0, x1, x2, x3)
5.95 –556.6665
f(x0, x1, x2, x3)
0.2 3.19000 –92.8333
f(x2, x3)
– 21.9
0.3 1.00000
Table
The second divided difference corresponding to x0, x1, x2 is given as,
f [x1, x2] – f [x0, x1]
f [x0, x1, x2] = x2 – x0
5.95 – (–33)
= 0.2 + 0.1
= 129.8333
f [x2, x3] – f [x1, x2]
f [x1, x2, x3] =
x3 – x1
= –21 .9 – 5.95
0.8 – 0.0 = – 92.8333
The third divided difference can be calculated as,
f [x1, x2, x3] – f [x0, x1, x2]
f [x0, x1, x2, x3] =
x3 – x0
= – 92.8333 – 129.8333
0.3 + 0.1
f [x0, x1, x2, x3] = – 556.6665.
According to Newton’s divided difference formula,
n
pn(x) = f [x0] + / f [x , x ...... x ] [(x – x ) ...... (x – x

0 1 k 0 k–1
)]
k=1
⇒ p3(x) = f [x0] + f[x0, x1] (x – x0) + f[x0, x1, x2] (x – x0) (x – x1) + f[x0, x1, x2, x3] (x – x0) (x – x1) (x – x2) ... (1)
p3(x) = 5.30000 + (– 33) (x – (– 0.1)) + 129.833 (x – (– 0.1)) (x – 0) – 556.66 (x – (– 0.1)) (x – 0) (x – 0.2)
= 5.3 – 33(x + 0.1) + 129.833(x + 0.1) (x) – 556.66 (x + 0.1) (x) (x – 0.2)
\ p3(x) = 5.3 – 33(x + 0.1) + 129.833(x) (x + 0.1) – 556.66 (x) (x + 0.1) (x – 0.2).
Q7. Use Newton’s forward difference formula to construct intepolating polynomial of degree one, two and
J –1 N
three for the following data. Approximate the specified value using each of the polynomials f KK 3 OO if
f(– 0.75) = – 0.07181250, f(– 0.5) = – 0.02475000, f(– 0.25) = 0.33493750, f(0) = 1.10100000. L P
Answer :
Given that,
x0 = – 0.75, x1 = – 0.5, x2 = – 0.25, x3 = 0
f(x0) = – 0.07181250, f(x1) = – 0.02475000, f(x2) = 0.33493750, f(x3) = 1.10100000
And h = 0.25
According to Newton’s forward difference formula,
First divided difference corresponding to x0, x1 is,
= Df(x0) = 1h [f(x0) – f(x0)]
= 0.125 [ – 0.02475000 – (– 0.07181250)]

⇒ D f(x0) = 0.18825.
The remaining first divided differences are shown in the table.
x f(x) First divided difference Second divided difference Third divided difference
– 0.75 – 0.07181250
Df(x0)
1.18825
D2f(x0)
– 0.5 – 0.02475000 1.002
Df(x1) D3f(x0)
1.43875 22
D2f(x1)
– 0.25 0.33493750 6.502
Df(x2)
3.06425
0.0 1.10100000
Table
Second divided difference corresponding to x0, x1, x2 is,
D2f(x0) = 1h [Df(x1) – Df(x0)]
= 0.125 [1.43875 – 1.18825]

D2f(x0) = 1.002
The remaining second order differences are shown in the table.
Third divided difference corresponding to x0, x1, x2, x3 is,
D3f(x0) = 1h [D2f(x1) – D2f(x0)]
= 0.125 [6.502 – 1.002]

⇒ D3f(x0) = 22.
According to Newton’s forward difference formula, interpolating polynomial of first degree is,
P1(x) = f(x0) + sh D f(x0) ... (1)
Where,
x0 = –31
–1
x–x – (– 0.75)
s = h 0 = 3 0.25 = 1.6666667
⇒ s = 1.6666667
Substituting the corresponding values in equation (1).
J N
P1 KK –31 OO = – 0.07181250 + (1.6666667) (0.25) (1.18825)
L P
J N
⇒ P1 KK –31 OO = 0.42322917
L P
Interpolating polynomial of second degree is,
s ( s – 1) h 2 2
P2(x) = P1(s) + 2! D f(x0) ... (2)
J N 1.6666667 (1.1666666 – 1) (0.25) 2
P2 KK –31 OO = 0.4232917 + 2 [1.1002]
L P
J N
P2 KK –31 OO = 0.432842
L P
Interpolating polynomial of third degree is,
s (s – 1) (s – 2) h 3 3
P3(x) = P2(s) + 3! D f(x0) ... (3)

J N 1.6666667 (1.16666667 – 1) (1.16666667 – 2) (0.25) 3
P3 KK –31 OO = 0.432842 + 6 (22)
L P
J N
P3 KK –31 OO = 0.4116229
L P
\ P1(0.43) = 0.4232917
P2(0.43) = 0.432842
P3(0.43) = 0.4116229
Q8. Use the Newton’s backward difference formula to construct interpolating polynomials of degree one,
J –1 N
two, three for f KK 3 OO if f( – 0.75) = – 0.07181250, f( – 0.5) = – 0.02475000, f( – 0.25) = 0.33493750, f(0) =
L P
1.10100000. Approximate the specified vaue using the polynomial.
Answer :
Given,
x0 = – 0.75, x1 = – 0.5, x2 = – 0.25, x3 = – 0, x4 = –31
f(x0) = – 0.07181250, f(x1) = – 0.02475000, f(x2) = 0.33493750, f(x3) = 1.10100000.

h = – 0.5 – (– 0.75)
⇒ h = 0.25
The first divided difference corresponding to xn and xn–1 is,
f [xn] – f [xn – 1]
∇f (xn) = h = 1.10100000 – 0.3349375
0 + 0.25 = 3.06425
The remaining first divided differences can be found in the same manner as shown in the table.
x f(x) ∇f ∇2f ∇3f

– 0.75 – 0.07181250
0.18825
– 0.5 – 0.02475000 5.002
1.43875 6
– 0.25 0.33493750 6.502 ∇ f(xn)
3
3.06425 ∇ f(xn)
2
∇f(xn)
0 1.10100000
Table
The second divided difference corresponding to xn, xn–1, xn–2 is,
4 f (xn) – 4 f (xn – 1)
∇2 f(xn) = h
= 3.064250.–251.43875
⇒ ∇2 f(xn) = 6.502
The remaining second divided differences are shown in the table, which are calculated in the same manner.
The third divided difference corresponding to xn, xn–1, xn–2, xn–3 is given as,
4 2 f (xn) – 4 2 f (xn–1)
∇3 f(xn) = h
= 6.5020.–255.002
\ ∇3 f(xn) = 6.
According to Newton’s backward difference formula,
The interpolating polynomial of degree one is given as,
P1(x) = f(xn) + hs ∇ f(xn) ... (1)
Where,
–1
s =
x – xn
= 3 –0 = –4
h 0.25 3

J N
⇒ P1(x) = 1.10100000 + 0.25 KK –34 OO (3.06425)
L P
JK – 4 NO
⇒ P1 K 3 O = 0.0795833
L P
JK –1 NO
\ f K 3 O = 0.0795833
L P
The interpolating polynomial of degree two is given as,
s (s + 1)
P2(x) = P1(x) + h2 ∇2 f(xn) 2! ... (2)
J N KKJ – 4 NOO JKK – 4 + 1NOO

⇒ P2 KK –31 OO = 0.0795833 + 0.406375
2
L P L 3 P L 3 P
J N
\ f KK –31 OO = 0.1698889
L P
The interpolating polynomial of degree three is given as,
s (s + 1) (s + 2)
P3(x) = P2(x) + h3 ∇3 f(xn) 3! ... (3)
s (s + 1) (s + 2)
⇒ P3(x) = P2(x) + (0.25)3 (6) 6
J N J N J NJ N J N
P3 KK –34 OO = P2 KK –34 OO + (0.25)3 KK –34 OO KK –34 + 1OO KK –34 + 2OO
L P L P L PL P L P
= 0.1698889 + 0.00462962
J N
⇒ f KK –31 OO = 0.1745185
L P

J N J N
\ f KK –31 OO = P1 KK –34 OO = 0.0795833
L P L P
JK –1 NO JK – 4 NO
f K 3 O = P2 K 3 O = 0.1698889
L P L P
JK –1 NO JK – 4 NO
f K 3 O = P3 K 3 O = 0.1745185.
L P L P
Q9. (a) Construct an approximating polynomial for the data given below using Hermite algorithm.
x f(x) f'(x)
0 1.00000 2.00000
0.5 2.71828 5.43656
(b) Using the polynomial constructed in (a), approximate f(0.43) and calculate the absolute error for
the function f(x) = e2x.
Answer :
Given data is,
x f(x) f ' (x)
0 1.00000 2.00000
0.5 2.71828 5.43656
The values are,
x0 = z0 = z1 = 0; x1 = z2 = z3 = 0.5
f(x0) = Q0,0 = Q1,0 = 1.00000
f(x1) = Q2,0 = Q3,0 = 2.71828
f '(x0) = Q1,1 = 2.00000
f '(x1) = Q3,1 = 5.43656
Q2,0 – Q1,0 2.71828 – 1.00000
Q2,1 = z2 – z1 = 0.5 – 0 = 3.43656
Q2,1 – Q1,1 3.43656 – 2.00000
Q2,2 = z2 – z0 = 0.5 – 0 = 2.87312
Q3,1 – Q2,1 5.43656 – 3.43656
Q3,2 = z3 – z1 = 0.5 – 0 =4
Q3,2 – Q2,2 4 – 2.87312
Q3,3 = z3 – z0 = 0.5 – 0 = 2.25376
z
z0 = 0 1.00000
2.00000
z1 = 0 1.00000 2.87312
3.43656 2.25376
z2 = 0.5 2.71828 4.00000
5.43656
z3 = 05 2.71828
The approximation of Hermite interpolating polynomial is given as,
H(x) = Q0,0 + Q1,1 (x – x0) + Q2,2 (x – x0)2 + Q3,3 (x – x0)2 (x – x1)

Substituting the corresponding values in above equation,
H(x) = 1.00000 + 2.00000 (x – 0) + 2.87312 (x – 0)2 + 2.25376 (x – 0)2 (x – 0.5)
= 1 + 2x + 2.87312 x2 + 2.25376 x2 (x – 0.5)
= 1 + 2x + 2.87312 x2 + 2.25376 x3 – 2.25376 x2 ( 0.5)
= 1 + 2x + 2.87312 x2 + 2.25376 x3 – 1.12688 x2
= 1 + 2x + 1.74624 x2 + 2.25376 x3
\ H(x) = 1 + 2x + 1.74624 x2 + 2.25376 x3
(b) Given function is,
f(x) = e2x ; x = 0.43
f(0.43) = e2(0.43) = 2.3631607
⇒ f(0.43) = 2.3631607
H(0.43) = 1 + 2(0.43) + 1.74624(0.43)2 + 2.25376 (0.43)3
= 2.3620695
Absolute error = |H(0.43) – f (0.43)|
= |2.3620695 – 2.3631607|
= | – 0.0010912 |
\ Absolute error = 0.0010912.
Q10. Let f(x) = 3xex – e2x, approximate f(1.03) by the Hermite interpolating polynomial of degree at most three
using x0 = 1 and x1 = 1.05.
Answer :
Given that,
f(x) = 3xex – e2x
⇒ f '(x) = 3[xex + ex (1)] – e2x(2)
= 3xex + 3ex – 2e2x.
And x0 = 1, x1 = 1.05
f(x0) = f(1) = 3(1)e(1) – e2(1) = 3e – e2 = 0.765789
f(x1) = f(1.05) = 3(1.05) e1.05 – e2(1.05)
⇒ f(x1) = 0.835431
f ' (x0) = 3(1)e(1) + 3e(1) – 2e2(1)
= 3e + 3e – 2e2.
⇒ f' (x0) = 1.531579
f'(x1) = 3(1.05) e(1.05) + 3e(1.05) – 2e2(1.05)

⇒ f'(x1) = 1.242214.
The table corresponding to the above values is,
x f(x) f'(x)
x0 = 1 f(x0) = 0.765789 f'(x0) = 1.531579
x1 = 1.05 f(x1) = 0.835431 f'(x1) = 1.242214
Table
According to Hermite method,
z0 = z1 = x0 = 1
z2 = z3 = x1 = 1.05
Q0,0 = f(x0) = 0.765789
Q1,0 = f(x0) = 0.765789
Q2,0 = f(x1) = 0.835431
Q3,0 = f(x1) = 0.835431
Q1,1 = f'(x0) = 1.531579
Q3,1 = f'(x1) = 1.242214
And
Qi,j – 1 – Qi – 1, j – 1
Qi,j = zi – zi – j
Q2,0 – Q1,0 0.835431 – 0.765789
Q2,1 = z2 – z1 = 1.05 – 1 = 1.39284
Q2,1 – Q1,1 1.39284 – 1.531579
Q2,2 = z2 – z0 = 1.05 – 1 = – 2.77478
Q3,1 – Q2,1 1.242214 – 1.39284
Q3,2 = z3 – z1 = 1.05 – 1 = – 3.01252
Q3,2 – Q2,2 – 3.01252 – (– 2.77478)
Q3,3 = z3 – z0 = 1.05 – 1 = – 4.7548
According to Hermite's interpolation formula,

H3(x) = Q0,0 + Q1,1 (x – x1) + Q2,2 (x – x0)2 + Q3,3 (x – x0)2 (x – x1) ... (1)
H3(x) = 0.765789 + 1.531579 (x – 1) + (– 2.77478) (x – 1)2 + (– 4.7548) (x – 1)2 (x – 1.05)
Substituting x = 1.03 in above equation,
H3(1.03) = 0.765789 + 1.531579 (1.03 – 1) – 2.77478 (1.03 – 1)2 – 4.7548 (1.03 – 1)2 (1.03 – 1.05)
⇒ H3(1.03) = 0.8093246
\ H3(1.03) = 0.8093246
Q11. Construct the natural cubic spline for the following data.
x f(x)
8.3 17.56492
8.6 18.50515
Answer :
Given,
x0 = 8.3, x1 = 8.6
f(x0) = 17.56492, f(x1) = 18.50515.
Let s(x) be a cubic polynomial and it is denoted by sj(x).
The equation for the natural cubic spline is,
sj(x) = aj + bj(x – xj) + cj(x – xj)2 + dj(x – xj)3 ... (1)
For j = 0
S0(x) = a0 + b0(x – x0) + c0(x – x0)2 + d0(x – x0)3
⇒ S0(x) = a0 + b0(x – 8.3) + c0(x – 8.3)2 + d0(x – 8.3)3 ... (2)
For n = 1
hi = xi + 1 – xj
Where,
i = 0, 1, 2 - - - -
⇒ h0 = x0 + 1 – x0
⇒ h0 = x 1 – x0
⇒ h0 = 8.6 – 8.3
\ h0 = 0.3
The vector equation is given as,
Ax = b
Where,
A is the matrix of (n + 1) × (n + 1)
b, x are vectors.
SR1 0WV
A = SSS WWW
0 1
T X2× 2
SR1 0WV
A = SSS WWW
0 1
T X
b = [0]
x = [c0]
RS1 0VW
SSS WWW [c0] = [0]
0 1
T X
⇒ c0 = 0
h
b0 = h1 (a1 – a0) – 30 (c1 + 2c0 )
0
= 01.3 (18.50515 – 17.56492) – 03.3 (2(0))

⇒ b0 = 3.1341
d0 = 31h (c1 – co)
0
= 3 (01.3) (– 0)
\ d0 = 0
S0(x) = 17.56492 + 3.1341(x – 8.3) + 0.00000 (x – 8.3)2 + 0.00000 (x – 8.3)3
\ S0(x) = 17.56492 + 3.1341(x – 8.3).
Q12. Construct the clamped cubic spline for the following data.
x f(x) f'(x)
8.3 17.56492 3.116256
8.6 18.50515 3.151762
Answer :
Given data,
x0 = 8.3, f(x0) = 17.56492, x1 = 8.6, f(x1) = 18.50515.
Cubic spline polynomial is given as,
S0(x) = a0 + b0(x – x0) + c0(x – x0)2 + d0(x – x0)3
S0(x) = a0 + b0(x – 8.3) + c0(x – 8.3)2 + d0(x – 8.3)2 ... (1)
The cubic spline must agree with the data at nodes,
(i) S(8.3) = f(8.3)
Substituting x = 8.3 in equation (1),
a0 + 0 + 0 + 0 = 17.56492
\ a0 = 17.56492 ... (2)
(ii) S(8.6) = f(8.6)
a0 + b0(8.6 – 8.3) + c0(8.6 – 8.3)2 + d0(8.6 – 8.3)3 = 18.50515
⇒ a0 + 0.3b0 + 0.09 c0 + 0.027 d0 = 18.50515
⇒ 17.56492 + 0.3b0 + 0.09 c0 + 0.027 d0 = 18.50515
\ 0.3b0 + 0.09 c0 + 0.027 d0 = 0.940230 ... (3)
Clamped Boundary Conditions
(ii) S'(x0) = f'(x0), (iv) S'(xn) = f'(xn)
(iii) S'(x0) = f'(x0)
S'(8.3) = f'(8.3)
S'(x) = b0 + 2 c0 (x – 8.3) + 3d0(x – 8.3)2
⇒ S'(8.3) = b0 + 2 c0 (0) + 3d0(0)
⇒ S'(8.3) = b0 = f ' (8.3)
\ b0 = 3.116256 ... (4)
(iv) S'(xn) = f'(xn)
⇒ S'(x1) = f'(x1)
⇒ S'(x) = b0 + 2 c0 (x – 8.3) + 3d0(x – 8.3)2
⇒ S'(8.6) = b0 + 2 c0 (8.6 – 8.3) + 3d0(8.6 – 8.3)2
\ S'(8.6) = b0 + 0.6 c0 + 0.27 d0
\ b0 + 0.6 c0 + 0.27 d0 = 3.151762 ... (5)
From equations (3) and (5),
0.3(3.116256) + 0.09 c0 + 0.027 d0 = 0.940230 ... (6)
3.116256 + 0.6 c0 + 0.27 d0 = 3.151762 ... (7)
Solving equations (6) and (7),
c0 = 0.06008666
d0 = – 0.00202222
S0(x) = 17.56492 + 3.116256 (x – 8.3) + 0.06008666(x – 8.3)2 – 0.0020222 (x – 8.3)3

Part-b
Essay Questions with Solutions
2.1 Interpolation and the Lagrange Polynomial

Q13. Write the Lagrange interpolating polynomial.
Answer :
Let P(x) be the polynomial and f(x) be the function
The linear nth Lagrange’s interpolating polynomial is given as,
P(x) = f(x0) Ln,0(x) + f(x1) Ln,1(x) + ......... + f(xn) Ln,n(x)

n
= / f (x ) L
k n, k (x)
k =0
Where,
k = 0, 1, 2, 3 ........ n.
(x – x ) (x – x ) ........ (x – x ) ...... (x – x )
Ln,k(x) = (x – x 0) (x – x1 ) ....... (x – xk + 1 ) ...... (x – xn )
1 0 k 1 k k+1 k n
n (x – x )
\ Ln,k(x) = p (x – xi )
i=0 k i
i!k
Q14. For the given functions f(x), let x0 = 0, x1 = 0.6, x2 = 0.9. Construct interpolation polynomials of degree
at most one and at most two to approximate f(0.45), and find absolute error.
(a) f(x) = 1+x
(b) f(x) = tanx
Answer : Model Paper-1, Q10(a)
Given function is,

f(x) = 1 + x
x0 = 0, x1 = 0.6, x2 = 0.9
f(x0) = 1 + 0 = 1
f(x1) = 1 + 0.6 = 1.264911
f(x2) = 1 + 0.9 = 1.378404
x–x
L0(x) = x – x1 = 0x –– 00..66
0 1
L0(x) = 0.60.–6 x
x–x
L1(x) = x – x0
1 0
= 0x.6––00
= 0x.6
The interpolating polynomial of degree at most one is given as,
P1(x) = L0(x) f(x0) + L1(x) f(x1) ... (1)

J N
P1(x) = KK 0.60.–6 x OO (1) + 0x.6 (1.264911)
L P
0.6 – x + 1.264911x
= 0.6
x + 0.6
= 0.264911 0.6
P1(x) = 0.441518x + 1
f(0.45) = 1 + 0.45 = 1.204153
P1(0.45) = 0.441518(0.45) + 1
\ P1(0.45) = 1.1986831
Absolute error is,
|f(0.45) – P1(0.45)| = |1.204153 – 1.1986831| = 0.0054699.
The interpolating polynomial of degree atmost two is given as,
P2(x) = L0(x) f(x0) + L1(x) f(x1) + L2(x) f(x2) ... (2)

Where,
(x – x ) (x – x ) (x – 0.6) (x – 0.9)
L0(x) = (x – x1) (x – x2 ) = (0 – 0.6) (0 – 0.9)
0 1 0 2
(x – 0.6) (x – 0.9)
= 0.54
(x – x ) (x – x )
L1(x) = (x – x0) (x – x2 )
1 0 1 2
(x – 0) (x – 0.9)
= (0.6 – 0) (0.6 – 0.9)
x (x – 0.9)
L1(x) = – 0.18
(x – x0) (x – x1)
L2(x) = (x – x ) (x – x )
2 0 2 1
(x – 0) (x – 0.6)
= (0.9 – 0) (0.9 – 0.6)
x (x – 0.6)
L2(x) = 0.27
(x – 0.6) (x – 0.9) x (x – 0.9) x (x – 0.6)
P2(x) = 0.54 (1) + – 0.18 (1.264911) + 0.27 (1.378404)
= 1.85185 [x2 – 1.5x + 0.54] – 7.027283 [x2 – 0.9x] + 5.1052000 [x2 – 0.6x]
= x2 [1.85185 – 7.027283 + 5.1052000 ] – x[– 2.77777 + 6.3245547 – 3.0631200] + 1
P2(x) = – 0.070233 x2 + 0.4836647x + 1
P2(0.45) = – 0.070233 (0.45)2 + 0.4836647 (0.45) + 1
P2(0.45) = 1.2034269
Absolute error is given as,
|f(0.45) – P2(0.45)| = |1.204153 – 1.2034269| = 0.0007261.
(b) f(x) = tan x
Given function is,
f(x) = tan x
x0 = 0, x1 = 0.6, x2 = 0.9
f(x0) = f(0) = tan 0 = 0
f(x1) = f(0.6) = tan (0.6) = 0.6841368
f(x2) = f(0.9) = tan (0.9) = 1.2601582
The interpolating polynomial of degree at most 1 is given as,
P1(x) = L0(x) f(x0) + L1(x) f(x1) ... (1)

x–x
L0(x) = x – x1
0 1
= 0x –– 00..66
⇒ L0(x) = 0.60.–6 x
x–x
L1(x) = x – x0
1 0
x–0
= 0.6 – 0
⇒ L1(x) = 0x.6

J N
P1(x) = KK 0.60.–6 x OO (0) + 0x.6 (0.6841368)
L P
= 0 + 1.1402280 x
= 1.1402280 x
\ P1(x) = 1.1402280 x
f(0.45) = tan(0.45) = 0.4830551
P1(0.45) = 1.1402280(0.45) = 0.5131026
Absolute error = | f(0.45) – P1(0.45)|
= |0.4830551 – 0.5131026|
= 0.0300475.
The interpolating polynomial of degree atmost two is given as,
P2(x) = L0(x) f(x0) + L1(x) f(x1) + L2(x) f(x2) ... (2)
(x – x ) (x – x ) (x – 0.6) (x – 0.9)
L0(x) = (x – x1) (x – x2 ) = (0 – 0.6) (0 – 0.9)
0 1 0 2
(x – 0.6) (x – 0.9)
= 0.54
(x – x ) (x – x ) (x – 0) (x – 0.9)
L1(x) = (x – x0) (x – x2 ) = (0.6 – 0) (0.6 – 0.9)
1 0 1 2
x (x – 0.9)
= – 0.18
(x – x ) (x – x )
L2(x) = (x – x0) (x – x1 )
2 0 2 1
(x – 0) (x – 0.6)
= (0.9 – 0) (0.9 – 0.6)
x (x – 0.6)
= 0.27
(x – 0.6) (x – 0.9) x (x – 0.9) x (x – 0.6)
P2(x) = 0.54 (0) + – 0.18 (0.6841368) + 0.27 (1.2601582)
= 0 – 3.8007600 x (x – 0.9) + 4.6672526 x (x – 0.6)
= – 3.8007600 x2 + 3.4206840 x + 4.6672526 x2 – 2.8003516x
= 0.8664926 x2 + 0.6203324 x
\ P2(x) = 0.8664926 x2 + 0.6203324 x
P2(0.45) = 0.8664926 (0.45)2 + 0.6203324 (0.45)
= 0.4546143
f (0.45) = tan(0.45) = 0.4830551
Absolute error = |f(0.45) – P2(0.45)|
= |0.4830551 – 0.4546143|
= |0.0284408|
\ Absolute error = 0.0284408
Q15. For the given functions f(x), let x0 = 1, x1 = 1.25, x2 = 1.6. Construct interpolation polynomials of degree
at most one and at most two to approximate f(1.4), and find absolute error, error bound.
(a) f(x) = e2x – x.
(b) f(x) = 3
x–1
Answer :
(a) f(x) = e2x – x
Given function is,
f(x) = e2x – x ... (1)
And
x0 = 1, x1 = 1.25, x2 = 1.6
f(x0) = f(1) = e2(1) –1 = 6.389056
f(x1) = f(1.25) = e2(1.25) –1 = 2.4903429
f(x2) = f(1.6) = e(1.6) –1 = 3.9530324
P1(x) = L0(x) f(x0) + f(x1) L1 (x) ... (2)
x–x
L0(x) = x – x1
0 1
= 1x –– 11..25
25
= x––01.25
.25
x–x
L1(x) = x – x0
1 0
= 1.x25– –1 1 = 0x.25
–1

J .25 NO Jx –1N
P1(x) = KK x––01.25 O (6.389056) + KK 0.25 OO (2.4903429)
L P L P
= –25.556224[x – 1.25] + 9.9613716 [x – 1]
= – 15.5948524 x + 31.94528 – 9.9613716
⇒ P1(x) = – 15.5948524x + 21.9839084
P1(1.4) = 0.15111504
Absolute error is,
|f(1.4) – P1(1.4)| = |15.0446467 – 0.15111504| = 14.89353166.
P2(x) = f(x0) L0(x) + f(x1) L1 (x) + f(x2) L2(x) ... (3)
Where,
(x – x ) (x – x )
L0(x) = (x – x1) (x – x2 )
0 1 0 2
(x – 1.25) (x – 1.6) (x – 1.25) (x – 1.6)

= (1 – 1.25) (1 – 1.6) = (– 0.25) (– 0.6)
(x – 1.25) (x – 1.6)
⇒ L0(x) = 0.15
(x – x ) (x – x )
L1(x) = (x – x0) (x – x2 )
1 0 1 2
(x – 1) (x – 1.6) (x – 1) (x – 1.6)
= (1.25 – 1) (1.25 – 1.6) = (0.25) (– 0.35)
(x – 1) (x – 1.6)
⇒ L1(x) = – 0.0875
(x – x ) (x – x )
L2(x) = (x – x0) (x – x1 )
2 0 2 1
(x – 1) (x – 1.25) (x – 1) (x – 1.25)
= (1.6 – 1) (1.6 – 1.25) = (0.6) (0.35)
(x – 1) (x – 1.25)
⇒ L2(x) = 0.21
RS (x – 1.25) (x – 1.6) VW RS (x – 1) (x – 1.6) VW RS (x – 1) (x – 1.25) VW
P2(x) = 6.389056 SS 0.15
WW + 2.4903429 SS W
– 0.0875 W + 3.9530324
SS
0.21
WW
T X T X T X
= 42.593706 [x2 – 2.85 x + 2] – 28.46106 [x2 – 2.6 x + 1.6] + 18.82396 [x2 – 2.25 x + 1.25]
= x2 [42.593706 – 28.46106 + 18.82396] + x[ – 121.39206 + 73.99875 – 42.35391]
+ 85.187412 – 45.537696 + 23.52995
P2(x) = 32.956606 x2 – 89.74722 x + 63.179666
⇒ P2(1.4) = 32.956606(1.4)2 – 89.74722(1.4) + 63.179666
= 2.12850576
|f(1.4) – P2(1.4)| = |15.0446467 – 2.12850576|
= 12.91614094.
The error bound formula of degree 1 is given as,
| f 2 (x (x)) |
Error bound = 2! |(x – x0) (x – x1)| ... (4)

f(x) = e – x. 2x
f'(x) = 2e2x – 1.
f2(x) = f''(x) = 4e2x
⇒ f 2(ξ(x)) = 4e2(ξ(x))
As (ξ(x)) is strictly increasing on [1, 1.25].
⇒ Max |f 2(ξ(x))|(1,1.25) = Max |4 e2(ξ(x))|(1,1.25)
= Max [|4 e2(1)|, |4 e2(1.25)|]
= Max [29,556224, 48.729975]
= 48.729975
ξ(x) = 1.25
2 (1.25)
Error bound = 4e 2! |(1.4 – 1) (1.4 – 1.25)|
^48.729975h
= | 2 (1.4 – 1) (1.4 – 1.25)|
= 1.4689925.
f 3 (x (x))
Error bound = 3! (x – x0) (x – x1) (x – x2) ... (5)
f(x) = e2x – x
f'(x) = 2e2x – 1
f''(x) = 4e2x
f'''(x) = 8e2x
⇒ f 3(ξ(x)) = 8e2(ξ(x))
As (ξ(x)) is strictly increasing function on [1, 1.6].
⇒ ξ(x) = 1.6
2 (1.6)
Error bound = 8e3! (x – 1) (x – 1.25) (x – 1.6)
At x = 1.4.
\ Error bound = 34 e3.2 (1.4 – 1) (1.4 – 1.25) (1.4 – 1.6)
= 0.3925204.
(b) f(x) = 3
x–1
Given function is,
f(x) = 3
x –1
= (x – 1)1/3 ... (1)
x0 = 1, x1 = 1.25, x2 = 1.6
f(x0) = f(1) = (1 – 1)1/3 = 0
f(x1) = f(1.25) = (1.25 –1)1/3 = 0.6299605
f(x2) = f(1.6) = (1.6 –1)1/3 = 0.8434326
P1(x) = L0(x) f(x0) + L1 (x) f(x1) ... (2)

x–x
L0(x) = x – x1 = 1x –– 11..25
25
= x––01.25
.25
0 1
x–x –1
L1(x) = x – x0 = 0x.25
1 0
J .25 NO Jx –1N
P1(x) = KK x––01.25 O (0) + KK 0.25 OO (0.6299605)
L P L P
= 2.519842 (x – 1)
⇒ P1(x) = 2.519842 x – 2.519842
P1(1.4) = 2.519842(1.4) – 2.519842
= 1.0079368
f(1.4) = 3
1.4 – 1 = 0.7368063
|f(1.4) – P1(1.4)| = |0.7368063 – 1.0079368|
= 0.2711305
P2(x) = f(x0) L0(x) + f(x1) L1 (x) + f(x2) L2(x) ... (3)

(x – x ) (x – x )
L0(x) = (x – x1) (x – x2 )
0 1 0 2
(x – 1.25) (x – 1.6) (x – 1.25) (x – 1.6)

= (1 – 1.25) (1 – 1.6) = 0.15
(x – x ) (x – x )
L1(x) = (x – x0) (x – x2 )
1 0 1 2
(x – 1) (x – 1.6)
= (1.25 – 1) (1.25 – 1.6)
(x – 1) (x – 1.6)
⇒ L1(x) = – 0.0875
(x – x ) (x – x )
L2(x) = (x – x0) (x – x1 )
2 0 2 1
(x – 1) (x – 1.25)
= (1.6 – 1) (1.6 – 1.25)
(x – 1) (x – 1.25)
⇒ L2(x) = 0.21
SR (x – 1) (x – 1.6) WV R V
SSS (x – 1) (x – 1.25) WWW
P2(x) = 0 + 0.6299605 SS – 0.0875 WW + 0.8434326 0.21
T X T X
= – 7.199548[(x – 1) (x – 1.6)] + 4.0163457 [(x – 1) (x – 1.25)]
= – 7.199548[x2 – 2.6x + 1.6] + 4.0163457 [x2 – 2.25x + 1.25]
= x2[– 7.199548 + 4.0163457] + x[18.7188248 – 9.0367778] – 11.5192768 + 5.0204321
P2(x) = – 3.1832023 x2 + 9.682047x – 6.4988447.
P2(1.4) = – 3.1832023(1.4) + 9.682047(1.4) – 6.4988447.
2
= 0.816944592.
|f(1.4) – P2(1.4)| = |0.73680629 – 0.816944592|
= 0.0801383
Error Bound
f 2 (x (x))
Error bound = 2! (x – x0) (x – x1) ... (4)
f(x) = (x – 1)1/3
f'(x) = 13 (x – 1)–2/3
f''(x) = –92 (x – 1)–5/3
f 2(ξ(x)) = –92 (ξ(x) –1)–5/3

Max |f 2(ξ(x))|(1,1.25)
⇒ Max | –92 (ξ(x) –1)–5/3|(1,1.25)
⇒ Max [| –92 (1 – 1)–5/3|, | –92 (1.25 – 1)–5/3|]

⇒ Max [0, 2.2398]
= 2.2398.
\ ξ(x) = 1.25.
–2 –5/3
Error bound = 9 (1.25 – 1) | (1.4 – 1) (1.4 –1.25) |
2!
= | – 1.1199298(1.4 – 1) (1.4 – 1.25)|
\ Error bound = 0.06719578.
f 3 (x (x))
f(x) = (x – 1)1/3
f'(x) = 13 (x – 1)–2/3
f''(x) = –92 (x – 1)–5/3
f '''(x) = 10
27 (x –1)
–8/3
f '''(ξ(x)) = 10
27 (ξ(x) –1)
–8/3

Max |f 2(ξ(x))|(1,1.6)
⇒ Max [| 10
27 (ξ(x) –1) |]
–8/3
⇒ Max [| 10 –8/3 10
27 (1 –1) |, | 27 (1.6 –1) |] = Max [0, 1.44621]
–8/3
\ ξ(x) = 1.6
10
27 (1.6 – 1) –8/3
= 3! (x – 1) (x – 1.25) (x – 1.6)
= |0.2410358 (1.4 – 1) (1.4 – 1.25) (1.4 – 1.6)|
\ Error bound = 0.00289243.
Q16. Use the following values and four digit rounding arithmetic to construct a third Lagrange polynomial
approximation to f(1.09) and the function being approximated as f(x) = log10(tanx). Find the bound for
the error in the approximation.
f(1.00) = 0.1924, f(1.05) = 0.2414, f(1.10) = 0.2933, f(1.15) = 0.3492
Answer :
Given data is,
f(1.00) = 0.1924; f(1.05) = 0.2414
f(1.10) = 0.2933; f(1.15) = 0.3492
The interpolating polynomial of degree 3 is given as,
P3(x) = L0(x) f(x0) + L1(x)f(x1) + L2(x) f(x2) + L3(x)f(x3) ... (1)
(x – x ) (x – x ) ( x – x )
L0(x) = (x – x 1) (x – x2 ) (x – 3x )
0 1 0 2 0 3
(x – 1.05) (x – 1.10) (x –1.15)

= (1.00 – 1.05) (1.00 – 1.10) (1.00 – 1.15)
(x – 1.05) (x – 1.10) (x –1.15)

= – 0.00075
(x – x ) (x – x ) (x – x )
L1(x) = (x – x 0) (x – x2 ) (x – x3 )
1 0 1 2 1 3
(x – 1.00) (x – 1.10) (x –1.15)

= (1.05 – 1.00) (1.05 – 1.10) (1.05 – 1.15)
(x – 1.00) (x – 1.10) (x –1.15)

= 0.00025
(x – x0) (x – x1) (x – x3)
L2(x) = (x – x ) (x – x ) (x – x )
2 0 2 1 2 3
(x – 1.00) (x – 1.05) (x –1.15)

= (1.10 – 1.00) (1.10 – 1.05) (1.10 – 1.15)
(x – 1.00) (x – 1.05) (x –1.15)

= – 0.00025
(x – x ) (x – x ) (x – x )
L3(x) = (x – x 0) (x – x1 ) (x – 2x )
3 0 3 1 3 2
(x – 1.00) (x – 1.05) (x –1.10)

= (1.15 – 1.00) (1.15 – 1.05) (1.15 – 1.10)
(x – 1.00) (x – 1.05) (x –1.10)

= 0.00075
(x – 1.05) (x – 1.10) (x –1.15) (x – 1.00) (x – 1.10) (x –1.15) (x – 1.00) (x – 1.05) (x –1.15)

P3(x) = – 0.00075 (0.1924) + 0.00025 (0.2414) + – 0.00025
(x – 1.00) (x – 1.05) (x –1.10)
(0.2933) + 0.00075 (0.3492)
= – 256.5333333 (x3 – 3.3x2+3.6275 x – 1.3283) + 965.6000 (x3 – 3.25x2 + 3.5150 x – 1.2650) – 1173.2000 (x3 – 3.2x2
+ 3.4075x – 1.2075) + 465.60 (x3 – 3.15 x2 + 3.3050 x – 1.1550)
= – 256.5333333 x3 + 846.5599999 x2 + 930.5746665 x + 340.7532266 + 965.60x3 – 3138.20 x2 + 3394.0840 x –
1221.4840 – 1173.2x3 + 3754.24x2 –3997.6790 x+1416.6390+465.60 x3 – 1466.64 x2 + 1538.8080 x – 537.7680
= 1.4667 x3 – 4.0400 x2 + 4.6383 x – 1.8598.
\ P3(x) = 1.4667 x3 – 4.04 x2 + 4.6383 x – 1.8598
P3(1.09) = 1.4667 (1.09)3 – 4.04 (1.09)2 + 4.6383(1.09) – 1.8598
⇒ P3(1.09) = 0.2954
f(1.09) = log10 (tan(1.09)) = 0.2826
Actual eror = |P(1.09) – f(1.09)|
= |0.2954 – 0.2826|
= |0.0128|
= 0.0128
Error Bound
f 4 (x (x))
Error bound = 4! (x – x0) (x – x1) (x – x2) (x – x3) ... (2)
f(x) = log10 (tanx)
ln (tan x)
f(x) = ln 10
f'(x) = ln110 tan

1
x .sec x
2
2
⇒ f'(x) = ln110 sec
tan x
x
Jd N Jd N
1 tan x KK dx (sec 2 x)OO – sec 2 x KK dx (tan x)OO
f''(x) = ln 10 L P L P
tan 2 x
RS 2 V
SS tan x.2 sec x (sec x. tan x) – sec x. sec x WWW
2
= ln110 S W
tan 2 x
T X
SRS 2 sec 2 x tan 2 x – sec 4 x WVW
= ln110 SS WW
tan 2 x
T X
SRS sec 4 x WV
= ln110 SS2 sec x – tan 2 x WWW
2
T X
SRS d d JK sec 4 x NOWV
f'''(x) = ln110 SS2. dx (sec x) – dx KK tan 2 x OOWWW
2
T L PX
RS RS VV
SS SS tan 2 x. d (sec 4 x) – sec 4 x d (tan 2 x) WWWWWW
= ln110 SS2 (2 sec x) . sec x tan x – SS
S S
dx dx WWWW
WW
(tan 2 x) 2
T T XX
SRS SR tan 2 x (4 sec3 x) . sec x tan x – sec 4 x (2 tan x) . sec 2 x WVWWVW
= ln110 SS4 sec 2 x tan x – SSS WWWW
tan 4 x
T T XX
SRS SRS 4 tan3 sec 4 x –2 tan x sec6 x WVWWVW
= ln110 SS4 sec x tan x – SS
2
WWWW
tan 4 x
T T XX
RS 4 6 V W
= ln110 SS4 sec 2 x tan x –
S
4 sec x 2
tan x + tan3 x WW
sec x W
T X
RS d RS sec 4 x VW RS sec6 x VWVW
fIV(x) = ln110 SS4. (sec x tan x) – 4. SS
S dx
2 d W S
dx T tan x WX + 2 SS tan3 x WWWW
WW
T T XX
SRS SRS d d WV
SS tan x. dx (sec x) – sec x. dx (tan x) WWW
4 4
SS RS 2 .d V
W
= ln110 SS4. SSsec x dx (tan x) + tan x. dx (sec x)WW – 4 SS
d 2
WW + 2
T X (tan x) 2
T T X
SRS 3 d d 3 W
VWV
SS tan x. dx (sec x) – sec x. dx (tan x) WWWWWW
6 6
SS WWWW
(tan3 x) 2
T XX
RS R x – sec 4 x. sec 2 x VWW +
SS4. 6sec 2 x. sec 2 x + tan x.2 sec x. sec x tan x@ – 4 SSS tan x.4 sec x. sec x tan
3
= ln110 S S WW 2
tan 2 x
T T X
SRS tan3 x.6 sec5 x. sec x tan x – sec6 x.3 tan 2 x. sec 2 x WVWWVW
SS WWWW
tan6 x
T XX
SRS R 4 sec 4 x tan 2 x – sec6 x WV SR 6 sec6 x tan 4 x – 3 tan 2 x sec8 x WVWV
SS4. 6sec x + 2 sec x tan x@ – 4 SSS
S
= ln110 4 2 2 WW + 2 SS
W S
WWWW
WW
tan 2 x tan6 x
T T X T XX
RS 4 2 6 6 4 2 8 VW
= ln110 SS4 sec 4 x + 8 sec 2 x tan 2 x –
S
16 sec x tan x +
4 sec x +
12 sec x tan x
–
6 tan x sec x WW
W
tan 2 x tan 2 x tan6 x tan6 x
T X
RS 6 6 8 V W
= ln110 SS4 sec x + 8 sec x tan x – 16 sec x +
4 2 2 4 4 sec x + 12 sec x
–
6 sec x W
S tan 2 x tan 2 x tan 4 x WW
T X
RS 6
x 6 sec x WW 8 V
= ln110 SS8 sec 2 x tan 2 x – 12 sec 4 x + 16 sec –
S tan 2 x tan 4 x WW
T X
RS 8 2 2
cos 4 x VWW
= ln110 SS 2 . sin x 12 16 cos
S cos x cos 2 x – cos 4 x + cos 4 x . sin 2 x – cos 4 x . sin 4 x WW
x 6
T X
RS 8 sin 2 x VW
= ln110 SS 12 16
S cos 2 x – cos 4 x + cos 4 x sin 2 x – cos 4 x sin 4 x WW
6 W
T X
SRS 1 SRS 6 WVWV
= ln110 16
SS cos 4 x SS8 sin x – 12 + sin 2 x – sin 4 x WWWWWW
2
T T XX
R
SS VW
f 4(ξ(x)) = ln110 1
S8 sin 2
( x ( x )) – 12 + 16
–
6 W
cos (x (x)) S 4
sin (x (x)) sin (x (x)) WW
2 4
T X
As (ξ(x)) is strictly increasing function on [1.00, 1.15],

1 16 6
Max| f 4(ξ(x))|(1.00, 1.15) = Max [8 sin 2 (x (x)) –12 + –
ln (10) cos 4 (x (x)) sin 2 (x (x)) sin 4 (x (x)) (1.00, 1.15)
= 81.50829256
Error bound = 81.50829256
24 (1.09 – 1.00) (1.09 – 1.05) (1.09 – 1.10) (1.09 – 1.15)
= 3.396178857 |0.000002160|
= 0.000007336
\ Error bound = 0.000007336
Q17. (a) Approximate the value of f(0.43) using the Lagrang's interpolating polynomials of degree one, two
and three if f(0) = 1, f(0.25) = 1.64872, f(0.5) = 2.71828, f(0.75) = 4.48169.
(b) Also find the error bound for the function f(x) = e2x and compare the bound to the actual error for
the cases n = 1, n = 2.
Answer :
Given data is,
f(0) = 1,
f(0.25) = 1.64872,
f(0.5) = 2.71828
f(0.75) = 4.48169
P1(x) = L0(x) f(x0) + L1 (x) f(x1) ... (1)

x–x
L0(x) = x – x1
0 1
= 0x –– 00..25
25
⇒ L0(x) = x––00.25
.25
x–x
L1(x) = x – x0
1 0
= 0.x25– –0 0
= 0.x25
P1(x) = x––00.25
.25
(1) + 0.x25 (1.64872)
= – 4(x – 0.25) + 6.59488 x
= – 4x + 1 + 6.59488 x
⇒ P1(x) = 2.5948800 x + 1
P2(x) = L0(x) f(x0) + L1(x)f(x1) + L2(x) f(x2) ... (2)
(x – x ) (x – x )
L0(x) = (x – x1) (x – x2 )
0 1 0 2
(x – 0.25) (x – 0.5) (x – 0.25) (x – 0.5)

= (0 – 0.25) (0 – 0.5) = 0.125
(x – x ) (x – x )
L1(x) = (x – x0) (x – x2 )
1 0 1 2
(x – 0) (x – 0.5) x (x – 0.5)
= (0.25 – 0) (0.25 – 0.5) = – 0.0625
(x – x ) (x – x )
L2(x) = (x – x0) (x – x1 )
2 0 2 1
(x – 0) (x – 0.25)
= (0.5 – 0) (0.5 – 0.25)
x (x – 0.25)
= 0.125
(x – 0.25) (x – 0.5) x (x – 0.5) x (x – 0.25)
P2(x) = 0.125 (1) + – 0.0625 (1.64872) + 0.125 (2.71828)
= 8(x2 – 0.25) (x – 0.5) – 26.37952x (x – 0.5) + 21.74624 x (x – 0.25)
= 8[x2 – 0.75x + 0.125] – 26.37952x2 + 13.18976x + 21.74624 x2 – 5.4365600 x
= 8x2 – 6x + 1 – 4.63328 x2 + 7.7532 x
= 3.36672 x2 + 1.7532 x + 1
\ P2(x) = 3.36672 x2 + 1.7532 x + 1
P3(x) = L0(x) f(x0) + L1(x)f(x1) + L2(x) f(x2) + L3(x)f(x3) ... (3)
(x – x ) (x – x ) ( x – x )
L0(x) = (x – x 1) (x – x2 ) (x – 3x )
0 1 0 2 0 3
(x – 0.25) (x – 0.5) (x – 0.75)

= (0 – 0.25) (0 – 0.5) (0 – 0.75)
(x – 0.25) (x – 0.5) (x – 0.75)

= – 0.09375
(x – x ) (x – x ) (x – x )
L1(x) = (x – x 0) (x – x2 ) (x – x3 )
1 0 1 2 1 3
(x – 0) (x – 0.5) (x – 0.75)
= (0.25 – 0) (0.25 – 0.5) (0.25 – 0.75)
x (x – 0.5) (x – 0.75)
= 0.03125
(x – x0) (x – x1) (x – x3)
L2(x) = (x – x ) (x – x ) (x – x )
2 0 2 1 2 3
(x – 0) (x – 0.25) (x – 0.75)
= (0.5 – 0) (0.5 – 0.25) (0.5 – 0.75)
x (x – 0.25) (x – 0.75)
= – 0.03125
(x – x0) (x – x1) (x – x2)
L3(x) = (x – x ) (x – x ) (x – x )
3 0 3 1 3 2
(x – 0) (x – 0.25) (x – 0.5)
= (0.75 – 0) (0.75 – 0.25) (0.75 – 0.5)
x (x – 0.25) (x – 0.5)
= 0.09375
(x – 0.25) (x – 0.5) (x – 0.75) x (x – 0.5) (x – 0.75) x (x – 0.25) (x – 0.75)
P3(x) = – 0.09375 (1) + 0.03125 (1.64872) + – 0.03125 (2.71828)
x (x – 0.25) (x – 0.5)
+ 0.09375 (4.48189)
= – 10.6666667 (x – 0.25) (x – 0.5) (x – 0.75) + 52.75904 x(x – 0.5) (x – 0.75) – 86.98496 x (x – 0.25)
(x – 0.75) + 47.8068267x (x – 0.25) (x – 0.5)
= – 10.6666667(x3 – 1.5 x2 + 0.6875 x – 0.09375) + 52.75904 (x3 – 1.25 x2 + 0.375 x)
– 86.98496 (x3 – x2 + 0.1875 x) + 47.8068267 (x3 – 0.75 x2 + 0.125x)
= – 10.6666667 x3 + 16 x2 – 7.3333334 x + 1 + 52.75904 x3 – 65.9488 x2 + 19.78464 x – 86.98496 x3
+ 86.98496 x2 – 16.30968 x + 47.8068267 x3 – 35.85512 x2 + 5.9758533 x
= 2.91424 x3 + 1.18104 x2 + 2.1174799 x + 1
\ P3(x) = 2.19424 x3 + 1.18104 x2 + 2.1174799 x + 1.
(b)
| f 2 (x (x)) |
Error bound = 2! |(x – x0) (x – x1)| ... (4)
f(x) = e 2x
f'(x) = 2e2x
f''(x) = f 2(x) = 4e2x
f 2(ξ(x)) = 4e2(ξ(x))
As (ξ(x)) is strictly increasing on [0, 0.25];
Max|f 2(ξ(x))|(0, 0.25) = |4e2(ξ(x))|(0, 0.25) = 6.5948851.
2! |(0.43 – 0) (0.43 – 0.25)|
= 6.5948851 2 |(0.43)(0.18)|
= (3.2974426) (0.0774)
= 0.2552221
\ At n = 1, Error bound = 0.2552221.
The error bound formula of degree 2 is given as
| f 3 (x (x)) |
Error bound = 3! |(x – x0) (x – x1) (x – x2)| ... (5)
f(x) = e2x
f'(x) = 2e2x
f 2(x) = f ''(x) = 4e2x
f 3(x) = f '''(x) = 8e2x
f 3(ξ(x)) = 8e2(ξ(x))
As (ξ(x)) is strictly increasing function on [0, 0.5]
Max|f 3(ξ(x))|(0, 0.5) = Max|8e2(ξ(x))|(0, 0.5) = 21.7462546
Error bound|x = 0.43 = 21.7462546
6 |(0.43 – 0) (0.43 – 0.25) (0.43 – 0.5)|
= (3.6243758) | – 0.0054180|
= (3.6243758) (0.0054180)
= 0.0196369
f(0.43) = e2(0.43) = 2.3631607
⇒ f(0.43) = 2.3631607
P1(0.43) = 2.59488 (0.43) + 1
⇒ P1(0.43) = 2.1157984
Absolute error|n = 1 = |f(0.43) – P1(0.43)|
= |2.3631607 – 2.1157984|
= 0.2473623
P2(0.43) = 3.36672(0.43)2 + 1.7532(0.43) + 1
⇒ P2(0.43) = 2.3763825
Absolute error|n = 2 = |f(0.43) – P2(0.43)|
= |2.3631607 – 2.3763825|
= | – 0.0132218|
n Error bound Absolute error

1 0.2552221 0.2473623
2 0.0196369 0.0132218
Q18. (a) Use appropriate Lagranges interpolating polynomials of degree one, two and three to approximate
f(0.18) if f(0.1) = – 0.29004986, f(0.2) = – 0.56079734, f(0.3) = –0.81401972, f(0.4) = – 1.0526302.
(b) Find the bound for the error for the cases n = 1 and n = 2 and compare with actual error, for the
function x2cosx – 3x.
Answer :
Given data is,
f(0.1) = – 0.29004986
f(0.2) = – 0.56079734
f(0.3) = – 0.81401972
f(0.4) = – 1.0526302
P1(x) = L0(x) f(x0) + L1(x) f(x1) ... (1)
x–x
L0(x) = x – x1 = 0x.1––00.2.2 = x––00.1.2
0 1
x–x
L1(x) = x – x0 = 0x.2––00.1.1 = x––00.1.1
1 0
P1(x) = x––00.1.2 (– 0.29004986) + x––00.1.1 (–0.56079734)
= 2.9004986 (x – 0.2) – 5.6079734(x – 0.1)

= 2.9004986 x – 0.5800997 – 5.6079734 x + 0.5607973
= – 2.7074748 x – 0.0193024
\ P1(x) = – 2.7074748 x – 0.0193024.
P2(x) = L0(x) f(x0) + L1(x) f(x1) + L2(x) f(x2) ... (2)
(x – x ) (x – x )
L0(x) = (x – x1) (x – x2 )
0 1 0 2
(x – 0.2) (x – 0.3) (x – 0.2) (x – 0.3)

= (0.1 – 0.2) (0.1 – 0.3) = 0.02
(x – x ) (x – x )
L1(x) = (x – x0) (x – x2 )
1 0 1 2
(x – 0.1) (x – 0.3) (x – 0.1) (x – 0.3)
= (0.2 – 0.1) (0.2 – 0.3) = – 0.01
(x – x ) (x – x )
L2(x) = (x – x0) (x – x1 )
2 0 2 1
(x – 0.1) (x – 0.2) (x – 0.1) (x – 0.2)

= (0.3 – 0.1) (0.3 – 0.2) = 0.02
(x – 0.2) (x – 0.3) (x – 0.1) (x – 0.3) (x – 0.1) (x – 0.2)
P2(x) = 0.02 (– 0.29004986) + – 0.02 ( – 0.56079734) + 0.02 ( – 0.81401972)
= – 14.5024930(x – 0.2) (x – 0.3) + 56.079734 (x – 0.1) (x – 0.3) – 40.7009860 (x – 0.1) (x – 0.2)

= – 14.5024930 (x2 – 0.5x + 0.06) + 56.079734 (x2 – 0.4x + 0.03) – 40.7009860 (x2 – 0.3x + 0.02)
= – 14.5024930 x2 + 7.2512465 x – 0.87014958 + 56.079734 x2 – 22.4318936 x + 1.68239202
– 40.7009860 x2 – 22.4318936 x + 1.68239202 – 40.7009860 x2 + 12.2102958 x – 0.81401972
= 0.876255 x2 – 2.9703513 x – 0.0017773
\ P2(x) = 0.876255 x2 – 2.9703513 x – 0.0017773
P3(x) = L0(x) f(x0) + L1(x) f(x1) + L2(x) f(x2) + L3(x) f(x3) ... (3)
(x – x ) (x – x ) ( x – x )
L0(x) = (x – x 1) (x – x2 ) (x – 3x )
0 1 0 2 0 3
(x – 0.2) (x – 0.3) (x – 0.4)

= (0.1 – 0.2) (0.1 – 0.3) (0.1 – 0.4)
(x – 0.2) (x – 0.3) (x – 0.4)

= – 0.006
(x – x0) (x – x2) (x – x3)
L1(x) = (x – x ) (x – x ) (x – x )
1 0 1 2 1 3
(x – 0.1) (x – 0.3) (x – 0.4)

= (0.2 – 0.1) (0.2 – 0.3) (0.2 – 0.4)
(x – 0.1) (x – 0.3) (x – 0.4)

= 0.002
(x – x ) (x – x ) (x – x )
L2(x) = (x – x 0) (x – x1 ) (x – 3x )
2 0 2 1 2 3
(x – 0.1) (x – 0.2) (x – 0.4)

= (0.3 – 0.1) (0.3 – 0.2) (0.3 – 0.4)
(x – 0.1) (x – 0.2) (x – 0.4)

= – 0.002
(x – x ) (x – x ) (x – x )
L3(x) = (x – x 0) (x – x1 ) (x – 2x )
3 0 3 1 3 2
(x – 0.1) (x – 0.2) (x – 0.3)

= (0.4 – 0.1) (0.4 – 0.2) (0.4 – 0.3)
(x – 0.1) (x – 0.2) (x – 0.3)

= 0.006
(x – 0.2) (x – 0.3) (x – 0.4) (x – 0.1) (x – 0.3) (x – 0.4)

P(x) = – 0.006 (– 0.29004986) + 0.002 (– 0.56079734) +
(x – 0.1) (x – 0.2) (x – 0.4) (x – 0.1) (x – 0.2) (x – 0.3)

– 0.002 ( – 0.81401972) + 0.006 (– 1.0526302)
= 48.3416433 (x – 0.2) (x – 0.3) (x – 0.4) – 280.39867 (x – 0.1) (x – 0.3) (x – 0.4)
+ 407.00986 (x – 0.1) (x – 0.2) (x – 0.4) – 175.4383667 (x – 0.1) (x – 0.2) (x – 0.3)
= 48.3416433 (x3 – 0.9x2 + 0.26x – 0.024) – 280.39867 (x3 – 0.8x2 + 0.19x – 0.012)
+ 407.00986 (x3 – 0.7x2 + 0.14x – 0.008) – 175.4383667 (x3 – 0.6x2 + 0.11x – 0.006)
= 48.3416433x – 43.5074790x + 12.5688273x – 1.1601994 – 280.39867 x + 224.3189360 x2
3 2 3
– 53.2757473x + 3.3647840 + 407.00986x3 – 284.9069020x2 + 56.9813804x – 3.2560789

– 175.4383667x3 + 105.2630200x2 – 19.2982203 x + 1.0526302
= – 0.4855334x3 + 1.1675750x2 – 3.0237599x + 0.0011359
\ P3(x) = – 0.4855334x3 + 1.1675750x2 – 3.0237599x + 0.0011359.
(b) The error bound formula of degree 1 is given as,
| f 2 (x (x)) |
Error bound = 2! |(x – x0) (x – x1)| ... (4)
f(x) = x2 cosx – 3x
f'(x) = x2 (– sin x) + cosx (2x) – 3
⇒ f'(x) = – x2 sin x + 2x cosx – 3
f''(x) = – [x2 (cosx) + sinx (2x)] + 2[x(– sinx) + cosx(1)] – 0
⇒ f''(x) = – x2 cosx – 2x sinx – 2xsinx + 2cosx – 0
⇒ f''(x) = – x2 cosx – 4x sinx + 2cosx
⇒ f''(ξ(x)) = – (ξ(x))2 cos(ξ(x)) – 4(ξ(x)) sin(ξ(x)) + 2cos(ξ(x))
As (ξ(x)) is strictly decreasing function on [0.1, 0.2]
Max|f''(ξ(x))|(0.1, 0.2) = Max| – (ξ(x))2 cos(ξ(x)) – 4(ξ(x)) sin(ξ(x)) + 2cos(ξ(x))| = 1.9401249
Error bound|x = 0.18 = 1.9401249
2 |(0.18 – 0.1) (0.18 – 0.2)|
= 0.9700625 | – 0.0016000|
= (0.9700625) (0.0016000)
= 0.0015521
\ Error bound (n = 1) = 0.0015521
f 3 (x (x))
f ''(x) = – x2 cos x – 4x sinx + 2cosx
f '''(x) = – (x2 (– sinx) + cosx (2x)) – 4(x cos x + sinx (1)) + 2(– sinx)
= x2 sinx – 2x cosx – 4x cosx – 4sinx – 2 sinx
= x2 sinx – 6x cosx – 6 sinx
f '''(ξ(x)) = (ξ(x))2 sin(ξ(x)) – 6(ξ(x)) cos(ξ(x)) – 6sin(ξ(x))
As (ξ(x)) is strictly decreasing on [0.1, 0.3]
Max|f '''(ξ(x))|(0.1, 0.3) = Max|(ξ(x))2 sin(ξ(x)) – 6(ξ(x)) cos(ξ(x)) – 6sin(ξ(x))|(0.1, 0.3)
= | – 3.4661301| – 3.4661301
3! |(0.18 – 0.1) (0.18 – 0.2) (0.18 – 0.3)|
= 0.5776884 | 0.0001920 |
= 0.5776884 (0.0001920)
= 0.0001109
\ Error bound (n = 2) = 0.0001109
f(0.18) = (0.18) cos (0.18) – 3(0.18)
2
⇒ f(0.18) = – 0.5081235
P1(0.18) = – 2.7074748 (0.18) – 0.0193024
⇒ P1(0.18) = – 0.5066479
= | – 0.5081235 – (– 0.5066479) |
= | – 0.0014756 |
= 0.0014756
P2(0.18) = 0.876255(0.18)2 – 2.9703513(0.18) – 0.0017773
⇒ P2(0.18) = – 0.5080499
= | – 0.5081235 – ( – 0.5080499) |
= | – 0.5081235 + 0.5080499 |
= | – 0.0000736 |
n Error bound Absolute error
1 0.0015521 0.0014756
2 0.0001109 0.0000736
2.2 Data Approximation and Neville’s Method

Q19. Explain the steps involved in interpolating the polynomial using Neville’s method.
Answer :
Step 1
Assign the values f(x0) = Q0,0, f(x1) = Q1,0 .......... f(xn) = Qn,0.
Step 2
Calculate the required approximations by using the formula,
(x – xi – j) Qi, j –1 – (x – xi) Qi–1, j –1
Qi,j = xi – xi – j
Where,
i = 1, 2 ......... n,
j = 1, 2 ......... i.
Step 3
The Recursive table is constructed as shown below.
xi Qi,0 Qi, 1 Qi, 2 Qi, n
x0 Q0,0
x1 Q1, 0 Q1, 1
x2 Q2, 0 Q2,1 Q2,2
xn Qn, 0 Qn, 1 Qn, 2 Qn, n
Table
Q20. Use Neville’s method to obtain the approximations for Lagrange interpolating polynomials of degrees
one, two and three to approximate f(8.4) if f(8.1) = 16.94410, f(8.3) = 17.56492, f(8.6) = 18.50515, f(8.7) =
18.82091.
Given data is,

f(8.1) = 16.94410, f(8.3) = 17.56492
f(8.6) = 18.50515, f(8.7) = 18.82091
The first degree approximation is given as,
(x – x0) Q1, 0 – (x – x1) Q0, 0
Q1,1(x) = x1 – x0
(x – 8.1) 17.56492 – (x – 8.3) 16.94410
⇒ Q1,1(x) = 8.3 – 8.1
+
= 17.56492 x – 142.27585200–.216.94410 x 140.6360300
= 0.6208200 0x.2– 1.6398220
\ Q1,1(x) = 0.6208200 0x.2– 1.6398220
0.6208200 (8.4) – 1.6398220

Q1,1(8.4) = 0.2
⇒ Q1,1(8.4) = 17.8753300
(x – x1) Q2, 0 – (x – x2) Q1, 0
Q2,1(x) = x2 – x1
(x – 8.3) 18.50515 – (x – 8.6) 17.56492
= 8.6 – 8.3
+
= 18.50515x – 153.59274500–.317.56492x 151.0583120
= 0.9402300x0.–3 2.5344330
\ Q2,1(x) = 0.9402300x0.–3 2.5344330
0.9402300 (8.4) – 2.5344330

Q2,1(8.4) = 0.3
⇒ Q2,1(8.4) = 17.878330
(x – x2) Q3, 0 – (x – x3) Q2, 0
Q3,1(x) = x3 – x2
(x – 8.6) (18.82091) – (x – 8.7) 18.50515
= 8.7 – 8.6
+
= 18.82091x – 161.85982600–.118.50515x 160.9948050
= 0.315760 x0.–10.8650210
0.315760 (8.4) – 0.8650210

Q3,1(8.4) = 0.1
⇒ Q3,1(8.4) = 17.8736300
The best linear approximation is considered as Q2,1, as 8.4 lies between 8.3 and 8.6.
The second degree approximation is given as,
(x – x0) Q2, 1 – (x – x2) Q1, 1
Q2,2(x) = x2 – x0
(x – 8.1) 17.878330 – (x – 8.6) 17.8753300
= 8.6 – 8.1
+
= 17.878330x – 144.81447300.–5 17.87533x 153.7278380
= 0.003 x –08.5.9133650
0.003 (8.4) – 8.9133650
Q2,2(8.4) = 0.5
⇒ Q2,2(8.4) = 17.8771300
(x – x1) Q3, 1 – (x – x3) Q2, 1
Q3,2(x) = x3 – x2
(x – 8.3) 17.8736300 – (x – 8.7) 17.878330
= 8.7 – 8.3
– 17.878330 + 155.5414710
= 17.8736300x – 148.3511290
0.4
+
= – 0.00470x0.47.1903420
– 0.00470 (8.4) + 7.1903420
Q3,2(8.4) = 0.4
⇒ Q3,2(8.4) = 17.8771550
The best quadratic approximation is considered as Q3,2 as 8.4 is near to 8.3.
The third degree approximation is given as,
(x – x0) Q3, 2 – (x – x3) Q2, 2
Q3,3(x) = x3 – x0
(x – 8.1) 17.8771550 – (x – 8.7) 17.8771300
= 8.7 – 8.1
17.877155x – 144.8049555 – 17.8771300 x + 155.5310310
= 0.6
0.000025x + 10.7260755
= 0.6
0.000025 (8.4) + 10.7260755
Q3,3(8.4) = 0.6
⇒ Q3,3(8.4) = 17.8771425
xi Qi,0 Qi,1 Qi,2 Qi,3

x0 8.1 16.94410 Q0,0
x1 8.3 17.56492 Q1,0 17.8733 Q1,1
x2 8.6 18.50515 Q2,0 17.87833 Q2,1 17.87713 Q3,2
x3 8.7 18.82091 Q3,0 17.87363 Q3,1 17.877155 Q3,2 17.8771425 Q3,3
one, two and three to approximate f(0.25) if f(0.1) = –0.62049958, f(0.2) = – 0.28398668, f(0.3) = 0.00660095,
f(0.4) = 0.24842440.
Answer :
Given data is,
f(0.1) = – 0.62049958, f(0.2) = – 0.28398668
f(0.3) = – 0.00660095, f(0.4) = – 0.24842440
(x – x0) Q1, 0 – (x – x1) Q0, 0
Q1,1(x) = x1 – x0
(x – 0.1) (– 0.28398668) – (x – 0.2) (– 0.62049958)
Q1,1(x) = 0.2 – 0.1
+ +
= – 0.28398668x 0.028398670.1 0.62049958x – 0.12409992
= 0.33651290 0x.1– 0.09570125
\ Q1,1(x) = 0.33651290 0x.1– 0.09570125

0.33651290 (0.25) – 0.09570125
Q1,1(0.25) = 0.1
⇒ Q1,1(0.25) = – 0.11573025
(x – x1) Q2, 0 – (x – x2) Q1, 0
Q2,1(x) = x2 – x1
(x – 0.2) 0.00660095 – (x – 0.3) (– 0.28398668)
= 0.3 – 0.2
0.00660095 x – 0.00132019 + 0.28398668 x – 0.08519600
= 0.1
0.29058763x – 0.08651619
= 0.1
0.29058763x – 0.08651619
\ Q2,1(x) = 0.1
0.29058763 (0.25) – 0.08651619
Q2,1(0.25) = 0.1
⇒ Q2,1(0.25) = – 0.13869283
(x – x2) Q3, 0 – (x – x3) Q2, 0
Q3,1(x) = x3 – x2
(x – 0.3) (0.24842440) – (x – 0.4) 0.00660095
= 0.4 – 0.3
+
= 0.24842440 x – 0.074527320–.10.00660095 x 0.00264038
= 0.24182345 0x .–1 0.07188694

0.24182345 (0.25) – 0.07188694
Q3,1(0.25) = 0.1
⇒ Q3,1(0.25) = – 0.11431078
The best linear approximation is considered as Q2,1 as 0.25 lies between 0.2 and 0.3.
(x – x0) Q2, 1 – (x – x2) Q1, 1
Q2,2(x) = x2 – x0
(x – 0.1) (–0.13869283) – (x – 0.3) (– 0.11573025)
= 0.3 – 0.1
+ +
= –0.13869283x 0.013869280.2 0.11573025x – 0.03471908
= – 0.022962580x.2– 0.02084980
– 0.02296258 (0.25) – 0.02084980

Q2,2(0.25) = 0.2
⇒ Q2,2(0.25) = – 0.13295223
(x – x1) Q3, 1 – (x – x3) Q2, 1
Q3,2(x) = x3 – x1
(x – 0.2) (– 0.11431078) – (x – 0.4) (– 0.13869283)
= 0.4 – 0.2
+ + 0.13869283x – 0.05547713
= – 0.11431078 x 0.02286216
0.2
= 0.02438205x0.–2 0.03261497
0.02438205 (0.25) – 0.03261497
Q3,2(0.25) = 0.2
⇒ Q3,2(0.25) = – 0.13259729
(x – x0) Q3, 2 – (x – x3) Q2, 2
Q3,3(x) = x3 – x0
(x – 0.1) (–0.13259729) – (x – 0.4) (– 0.13295223)
= 0.4 – 0.1
+ + 0.13295223x – 0.05318089
= – 0.13259729x 0.01325973
0.3
= 0.00035494 0x.3– 0.03992116

0.00035494 (0.25) – 0.03992116
Q3,3(0.25) = 0.3
⇒ Q3,3(0.25) = – 0.13277475

0.1 – 0.62049958
0.2 – 0.28398668 – 0.11573025
0.3 0.00660095 – 0.13869283 – 0.13295223
0.4 0.24842440 – 0.11431078 – 0.13259729 – 0.13277475
one, two and three to approximate f(0.43) if f(0) = 1, f(0.25) = 1.64872, f(0.5) = 2.71828, f(0.75) = 4.48169.
Answer :
Given data is,
f(0) = 1, f(0.25) = 1.64872
f(0.5) = 2.71828, f(0.75) = 4.48169
(x – x0) Q1, 0 – (x – x1) Q0, 0
Q1,1(x) = x1 – x0
(x – 0) (1.64872) – (x – 0.25) (1)
Q1,1(x) = 0.25 – 0
0 – x + 0.25
= 1.64872x 0–.25
x + 0.25
= 0.64872 0.25
x + 0.25
\ Q1,1(x) = 0.64872
0.25
0.64872 (0.43) + 0.25
Q1,1(0.43) = 0.25
⇒ Q1,1(0.43) = 2.11579840
(x – x1) Q2, 0 – (x – x2) Q1, 0
Q2,1(x) = x2 – x1
(x – 0.25) 2.71828 – (x – 0.5) 1.64872
= 0.5 – 0.25
+
= 2.71828 x – 0.679570.–251.64872 x 0.82436
+
= 1.069560x.250.14479
+
\ Q2,1(x) = 1.069560x.250.14479
1.06956 (0.43) + 0.14479

Q2,1(0.43) = 0.25
⇒ Q2,1(0.43) = 2.41880320
(x – x2) Q3, 0 – (x – x3) Q2, 0
Q3,1(x) = x3 – x2
(x – 0.5) (4.48169) – (x – 0.75) 2.71828
= 0.75 – 0.5
– 2.71828 x + 2.03871
= 4.48169 x – 2.240845
0.25
= 1.76341x0.–250.202135
1.76341 (0.43) – 0.202135

Q3,1(0.43) = 0.25
⇒ Q3,1(0.43) = 2.22452520
The best linear approximation is considered as Q2,1 as 0.43 lies between 0.25 and 0.5.
(x – x0) Q2, 1 – (x – x2) Q1, 1
Q2,2(x) = x2 – x0
(x – 0) 2.41880320 – (x – 0.5) 2.11579840
= 0.5 – 0
2.41880320x – 0 – 2.11579840 x + 1.05789920
= 0.5
0.30300480 x + 1.05789920
= 0.5
0.30300480 (0.43) + 1.05789920
Q2,2(0.43) = 0.5
⇒ Q2,2(0.43) = 2.37638253
(x – x1) Q3, 1 – (x – x3) Q2, 1
Q3,2 (x) = x3 – x1
(x – 0.25) 2.22452520 – (x – 0.75) 2.41880320
= 0.75 – 0.25
+
= 2.22452520 x – 0.556131300–.52.41880320 x 1.81410240
+
= – 0.194278 x0.5 1.25797110
– 0.194278 (0.43) + 1.25797110
Q3,2(0.43) = 0.5
⇒ Q3,2(0.43) = 2.34886312
(x – x0) Q3, 2 – (x – x3) Q2, 2
Q3,3(x) = x3 – x0
(x – 0) (2.34886312) – (x – 0.75) 2.37638253
= 0.75 – 0
x + 1.78228690
= 2.34886312 x – 0 – 2.037638253
.75
+
= – 0.027519410.x75 1.78228690
– 0.02751941 (0.43) + 1.78228690

Q3,3(0.43) = 0.75
⇒ Q3,3(0.43) = 2.36060474

0 1
0.25 1.64872 2.11579840
0.5 2.71828 2.41880320 2.37638253
0.75 4.48169 2.22452520 2.34886312 2.36060474
2.3 Divided Differences
Q23. Write the procedure to construct divided difference table.

Answer :
Divided Difference Table
Step 1
The first divided difference of f(x) with respect to xi and xi+1 is given as,
f [xi + 1] – f [xi]
f[xi, xi+1] = xi + 1 – xi
Where,
i = 0, 1, 2 ......
Step 2
The second divided difference of f(x) with respect to xi, xi+1 and xi+2 is given as,
f [xi + 1, xi + 2] – f [xi, xi + 1]
f[xi, xi+1, xi+2] = xi + 2 – xi
Step 3
The kth divided difference of f(x) with respect to xi, xi+1, xi+2 - - -, xi+k is given as,
f [xi + 1, xi + 2 - --- xi + k] – f [xi, xi + 1 - --- , xi + k – 1]
f[xi, xi+1, ........ xi + k – 1, xi + k] = xi + k – xi
Step 4
The process is continued until the desired divided difference is obtained.
Step 5
The divided difference table can be tabulated as shown below.
x f(x) First divided difference Second divided difference - - - - - kth divided difference
x0 f[x0]
f[x0, x1]
f [x1 ] – f [x 0 ] f[x0, x1, x2]
x1 – x 0
x1 f[x1] f [x1, x 0] – f [x 0, x1] f[xi, xi + 1, - - - xi + k]
x 2 – x0
f[x1, x2] f[xi+1, xi + 2, - - -, xi + k]
f [x 2 ] – f [x1 ] h – f [x i , x i + 1 - - - - x i + k – 1 ]
x 2 – x1 xi + k – xi
x2 f[x2] h f[xn – 2, xn – 1, xn]
h h f[xn – 1, xn] f [x n – 1 , x n ] – f [x n – 2 , x n – 1 ]
xn – xn – 2
f [x n ] – f [x n – 1 ]
xn – xn – 1
xn f[xn]
Table
Step 6
The divided difference interpolating polynomial is given as,
Pn(x) = f[x0] + f[x0, x1] (x – x0) + f[x0, x1, x2] (x – x0) (x – x1) + f[x0, x1, x2, x3] (x – x0) (x – x1) (x – x2) + - - - - - +
f[x0,- - xn] (x – x0) (x – x1) - - - - - - - (x – xn – 1).
Q24. Write the steps involved in order to construct forward difference table.
Answer :
Forward Difference
Step 1
The step-size (h) for the forward difference table can be calculated as,
h = xi + 1 – xi.
Where,
i = 0, 1, 2 .......
Step 2
The first divided difference of ‘f’ with respect to xi and xi + 1 is given as,
f [xi + 1] – f [xi]
f[xi, xi + 1] = xi + 1 – xi
Where,
i = 0, 1,........
Step 3
The second divided difference of f with respect to xi, xi + 1, xi + 2 is given as,
f [xi + 1, xi + 2] – f [xi, xi + 1]
f[xi, xi + 1, xi + 2] = xi + 21 – xi
Step 4
The process is continued until the desired nth divided difference is obtained.
Step 5
The forward difference table can be tabulated as shown below.
Value Value First difference Df Second difference D2f - - - - nth difference - - - - - Dnf
of x of f(x)
x0 f(x0)
f (x1) – f (x 0)
Df0 = h
x1 f(x1) D f1 – D f0
D2 f0 = h
f (x 2) – f (x1) D n–1 f1 – D n–1 f0
Df1 = h - - - Dnf0 = h
x2 f(x2) D f2 – D f1
D2 f1 = h
x3 f[x3] f (x 3) – f (x 2)
Df2 = h
xn f(xn)
Table
Step 6
The forward interpolating polynomial can be given as,
Pn[x] = f[x0] + hs D f[x0] + 2h! s(s + 1) D2 f[x0] + - - - - - -.

2

Q25. Explain the procedure to construct backward difference table.

Answer :
The procedure to construct backward difference table is shown below.
Step 1
The step-size (h) for the backward difference table can be calculated as,
h = xi+1 – xi
Where,
i = 0, 1, 2 . . . .. n – 1.
Step 2
The first divided difference of ‘f’ with respect to xi and xi + 1 is given as,
f [xi + 1] – f [xi]
f[xi, xi + 1] = xi + 1 – xi
Where,
i = 0, 1, . . . . . .
Step 3
The second divided difference of f with respect to xi, xi + 1, xi + 2 is given as,
f [xi + 1, xi + 2] – f [xi, xi + 1]
f[xi, xi + 1, xi + 2] = xi + 1 – xi
Step 4
The process is continued until the desired nth divided difference is obtained.
Step 5
The backward difference table can be tabulated as shown below.
Value Value First difference (∇f) Second difference nth difference (∇nf)
of x of f(x) (∇2f)
x0 f(x0)
f (x1) – f (x 0)
∇f1 = h
x1 f(x1) 4 f2 – 4 f1
∇ 2 f2 = h
f (x 2) – f (x1)
∇f1 = h
x2 f(x2)
(4 n–1 fn – 4 n–1 fn – 1)
∇nfn = h
xn–1 f(xn – 1) 4 fn – 4 fn–1

∇ 2 fn = h
f (x n) – f (x n–1)
∇fn = h
xn f(xn)
Table
Step 6
The backward interpolating polynomial is given as,
Pn[x] = f[xn] + hs ∇ f[xn] + 2h! s(s + 1) ∇2 f[xn] + 3h! s(s + 1) (s + 2) ∇3 f[xn] + - - - - - -.

2 2

Where,
x – xn
s = h
Q26. Using Newton’s divided-difference formula, construct interpolating polynomials of degree one, two
and three for the given data.
f(8.4) if f(8.1) = 16.94410, f(8.3) = 17.56492, f(8.6) = 18.50515, f(8.7) = 18.82091.
Approximate the specified value using the polynomial.
Answer :
Given,
x0 = 8.1, x1 = 8.3, x2 = 8.6, x3 = 8.7, x4 = 8.4.
f(x0) = 16.94410, f(x1) = 17.56492, f(x2) = 18.50515, f(x3) = 18.82091.
The first divided difference corresponding to x0 and x1 can be given as,
f [x1] – f [x0]
f[x0, x1] = x1 – x0
= 17.56492 – 16.94410
8.3 – 8.1
\ f[x0, x1] = 3.1041.
The remaining first divided differences are found in the same manner.
i xi f(xi) f(xi – 1, xi) f(xi –2, xi –1, xi) f(xi–3, xi–2, xi –1, xi)
0 8.1 16.94410 f(x0, x1)
3.1041
1 8.3 17.56492 f(x0, x1, x2)
f(x1, x2) 0.03566 f(x0, x1, x2, x3)
3.12193 0.0891916
2 8.6 18.50515 f(x1, x2, x3)
f(x2, x3) 0.089175
3.1576
3 8.7 18.82091
The second divided difference corresponding to x0, x1, x2 is given as,

f [x1, x2] – f [x0, x1]
f[x0 x1, x2] = x2 – x0 = 3.12193 – 3.1041
8.6 – 8.1
= 0.03566.
The remaining second divided differences are shown in the fifth column of the table. The third divided difference
corresponding to x0, x1, x2, x3 is given as,
f [x1, x2, x3] – f [x0, x1, x2]
f[x0, x1, x2, x3] = x3 – x0 = 0.089175 – 0.03566
8.7 – 8.1
f[x0, x1, x2, x3] = 0.0891916
According to Newton’s divided difference formula,
p1(x) = f(x0) + f(x0, x1) (x – x0) ... (1)
p1(x) = 16.94410 + 3.1041 (x – 8.1)
p1(8.4) = 16.94410 + 3.1041 (8.4 – 8.1)
\ p1(8.4) = 17.87533
p2(x) = f1(x) + f(x0, x1, x2) (x – x0) (x – x1) ... (2)
p2(x) = 16.94410 + 3.1041 (x – 8.1) + 0.03566 (x – 8.1) (x – 8.3)
⇒ p2(8.4) = 16.94410 + 3.1041 (8.4 –8.1) + 0.03566 (8.4 –8.1) (8.4 – 8.3)
\ p2(8.4) = 17.8763998.
Q27. Use Newton’s forward difference formula to constust interpolating polynomials of degree one, two and
three for the following data.
f(0.25) if f(0.1) = – 0.62049958, f(0.2) = – 0.28398668.
f(0.3) = 0.00660095, f(0.4) = 0.24842440. Approximate the specified value using each of the polynomials.
Given,
x0 = 0.1, x1 = 0.2, x2 = 0.3, x4 = 0.4.
f(x0) = – 0.62049958, f(x1) = – 0.28398668, f(x3) = 0.00660095, f(x4) = 0.24842440.
According to Newton's divided difference formula,
The first divided difference corresponding to x0, x1 is,
f[x0, x1] = Df(x0) = 1h [f(x1) – f(x0)] = 1h Df(x0)
Df(x0) = f[x0, x1] = 01.1 [– 0.28495668 – (– 0.62049958)]
⇒ Df(x0) = 3.365129.
The remaining first divided difference are found in same manner.
i xi f(xi) f(xi – 1, xi) f(xi –2, xi –1, xi) f(xi–3, xi–2, xi –1, xi)
0 0.1 – 0.62049958
Df(x0)
3.365129
D2 f(x0)
1 0.2 – 0.28398668 – 4.592527
Df(x1) D3f(x0)
2.9058763 – 2.83891
2 0.3 0.00660095 – 4.876418
Df(x2) D2f(x1)
2.4182345
3 0.4 0.24842440
Table
The second divided difference corresponding to x0, x1, x2 is,
Df (x1) – Df (x0)
D2 f(x0) = f[x0, x1, x2] =
h
= 2.9058763 – 3.365129
0.3 – 0.1
⇒ f[x0, x1, x2] = D2 f(x0) = – 4.592527
The remaining second divided differences are shown in the table.
D 2 f (x1) – D 2 f (x0)
D3 f(x0) = f[x0, x1, x2, x3] = h
– 2.4382209 – (– 2.2962635)
= 0.3 – 0.2
⇒ f[x0, x1, x2, x3] = D3 f(x0) = – 2.83891
According to Newton’s forward difference formula, interpolation of first degree polynomial is,
p1(x) = f(x0) + s.h. D f(x0) ... (1)
Where,
h = 0.1
x–x
s = h 0 = 0.250.–1 0.1 = 1.5
p1(0.25) = – 0.62049958 + (1.5) (0.1) [3.365129]
⇒ p1(0.25) = – 0.1157302.
Interpolation of second degree polynomial is,
s (s – 1) h 2 [D 2 f (x0)]
p2(x) = f1(s) + 2! ... (2)
(1.5) (1.5 – 1) (0.1) 2 [– 4.592527]
p2(0.25) = – 0.1157302 + 2!
⇒ p2(0.25) = – 0.1329522
Interpolation of third degree polynomial is,
s (s – 1) (s – 2) (h 3 [D 3 f (x0)])
p3(0.25) = p2(s) + 3!
(1.5) (1.5 – 1) (1.5 – 2) (0.1) 3
= – 0.1329522 + 6 [–2.83891]
p3(0.25) = – 0.13277474
\ p1(0.25) = – 0.1157302
p2(0.25) = – 0.1329522
p3(0.25) = – 0.13277474.
Q28. Use the Newton backward difference formula to construct interpolating polynomials of degree one,
two, three for the data f(0.25) if f(0.1) = – 0.62049958, f(0.2) = – 0.28398668, f(0.3) = 0.00660095, f(0.4) =
0.24842440. Approximate the specified value using the polynomial.
Answer :
Given,
x0 = 0.1, x1 = 0.2, x2 = 0.3, x3 = 0.4, x4 = 0.25
f(x0) = – 0.62049958, f(x1) = – 0.28398668, f(x2) = 0.00660095, f(x3) = 0.24842440.
h = 0.1
The first divided difference corresponding to xn and xn – 1 is,
f [xn] – f [xn – 1]
∇f(xn) =
h
= 0.248424400–.10.00660095
= 2.4182345.
x f(x) Newton’s First Second Backward Third Backward

Backward Divided Divided Difference Divided Difference
Difference (∇f) (∇2f) (∇3f)
0.1 – 0.62049958
3.365129
∇f(xn – 2)
0.2 – 0.28398668 – 4.592527
2.9058763 ∇2f(xn – 1) – 2.83891
∇f(xn – 1)
0.3 0.00660095 – 4.876418 ∇3f(xn)
2.4182345 ∇2f(xn)
∇f(xn)
0.4 0.24842440
f(xn)
Table
The second divided difference corresponding to xn, xn–1, xn–2 is given as,
4 f (xn) – 4 f (xn–1)
∇2 f(xn) = h = 0.241823450.–1 2.9058763
\ ∇2 f(xn) = – 4.876418.
The remaining second divided differences are shown in the table.
The third divided difference corresponding to xn, xn–1, xn–2, xn–3 is given as,
4 2 f (xn) – 4 2 f (xn–1) – 4.876418 – (– 4.592527)
∇3 f(xn) = h = 0.1
∇3 f(xn) = – 2.83891.
Newton’s backward difference formula is given as,
n JK –sNO
pn(x) = f(xn) + / (–1) k KK OO ∇ kf(xn) ... (1)
k=1
Lk P
p1(x) = f[xn] + h.s ∇ f(xn) ... (2)
⇒ p1(x) = 0.24842440 + s(0.24182345)
Where,
x – xn
s = h = 0.250.–1 0.4 = – 1.5
\ p1( – 1.5) = 0.24842440 + (– 1.5) (0.24182345)
f(0.25) = 0.24842440 – 0.36273517
\ f(0.25) = – 0.11431077.
s (s + 1)
p2(x) = p1(s) + h2 ∇2 f(xn) 2! ... (3)
J s (s + 1) N
p2(x) = 0.24842440 + 0.24182345 s + (0.1)2 ( – 4.876418) KK 2! OO
L P
J – 1.5 (– 1.5 + 1) ON
⇒ p2 (–1.5) = 0.24842440 + 0.24182345 (– 1.5) – 0.04876418 KK 2
O
L P
= 0.2484244 – 0.362735175 – 0.01828656
f(0.25) = p2 (–1.5) = – 0.132597335.
The interpolating polynomial of degree three is given as,
s (s + 1) (s + 2)
p3(x) = p2(s) + h3 ∇3 f(xn) 3! ... (4)
s (s + 1) (s + 2)
p3(s) = p2(s) + (0.1)3 (2.83891) 6
Substituting s = –1.5 in the above equation,
J (– 1.5) (–1.5 + 1) (–1.5 + 2) NO
⇒ p3 (–1.5) = p2 (–1.5) – 0.002838 KK 6
O
L P
⇒ p3 (–1.5) = – 0.132597335 – 0.002838 (0.0625)
\ f (0.25) = – 0.13277471.
\ p1 (–1.5) = f(0.25) = – 0.11431077
p2 (–1.5) = f(0.25) = – 0.132597335
p3 (–1.5) = f(0.25) = – 0.13277471

Q29. Approximate f(0.43) by Stirlings formula using the following data.
x 0.0 0.2 0.4 0.6 0.8

f(x) 1.00000 1.22140 1.49182 1.82212 2.22554
Answer :
Given,
x0 = 0.0, x1 = 0.2, x2 = 0.4, x3 = 0.6, x4 = 0.8
f(x0) = 1.00000, f(x1) = 1.22140, f(x2) = 1.49182, f(x3) = 1.82212, f(x4) = 2.22554.
First divided difference corresponding to x0, x1 is,
f [x1] – f [x0]
f[x0, x1] = x1 – x2
⇒ f[x0, x1] = 1.22140

0.2 – 0
–1
⇒ f[x0, x1] = 1.107.

The remaining values of first divided differences are shown in the table.
x f(x) First Divided Second Divided Third Divided Fourth Divided

Difference Difference Difference Difference
0 1.00000
1.107
0.2 1.22140
0.61275
f[x–1, x0] 0.22625
1.3521 f[x–2, x–1, x0, x1]
0.4 1.49182 0.7485 0.061979
f[x0] f[x–1, x0, x1] f[x–2, x–1, x0, x1, x2]
1.6515 0.275833
f[x0, x1]
0.6 1.82212 0.914 f[x–1, x0, x1, x2]
2.0171
0.8 2.22554
Table
Second divided difference corresponding to x0, x1, x2 is,
f [x0, x2] – f [x1, x0]
f [x0, x1, x2] = x2 – x0 = 1.35210.–41.107
⇒ f[x0, x1, x2] = 0.61275.
Remaining values of the second divided difference are shown in the table.
f [x1, x2, x3] – f [x2, x1, x0]
f [x0, x1, x2, x3] = x3 – x0 = 0.61275 – 0.7485
0.6 – 0
⇒ f [x0, x1, x2, x3] = 0.22625.
Remaining values of third divided difference are shown in the table.
Fourth divided difference corresponding to x0, x1, x2, x3, x4 is,
f [x1, x2, x3, x4] – f [x3, x2, x1, x0]
f[x0, x1, x2, x3, x4] = x4 – x0
⇒ f[x0, x1, x2, x3, x4] = 0.275833 – 0.22625

0.8 – 0
⇒ f[x0, x1, x2, x3, x4] = 0.061979
According to Stirling’s formula,

s ( s 2 – 1) h 3
pn(x) = f[x0] + sh2 [f[x–1
, x0
] + f[x0
, x1
]] + s h
2 2
f[x–1
, x0
, x1
] + 2 (f[x–2, x–1, x0, x1] + f(x–1, x0, x1, x2)]
+ s2(s2 – 1)h4 (f(x–2, x–1, x0, x1, x2) + . . .. . . ... (1)
Where,
x = 0.43
x – x0
s = h = 0.430.–2 0.4 = 0.15
⇒ s = 0.15.
(0.15) (0.2) (0.15) ((0.15) 2 –1) (0.2) 3
p(0.43) = 1.49182 + 2 [1.3521 + 1.6515] + (0.15]2
(0.2) 2
[0.7485] + 2
(0.22615 + 0.275833) + (0.15)2 (0.15)2 – 1) (0.2) + (0.061979)
⇒ p(0.43) = 1.537251
\ p(0.43) = 1.53725.
2.4 Hermite Interpolation
Q30. Explain the steps involved in approximating hermite interplating polynomial.

Answer :
Step 1
Assign the values,
x0 = z0 = z1, x1 = z2 = z3, ....... xn = z2n = z2n+1
f(x0) = Q0, 0 = Q1, 0, f(x1) = Q2, 0 = Q3, 0 ....... f(xn) = z2n0 = z2n+1, 0
f ' (x0) = Q1, 1, f ' (x1) = Q3, 1 , ....... f ' (xn) = Q2n+1, 1
Step 2
Calculate the required approximations by using the formula,
Qi,j – 1 – Qi–1, j –1
Qi, j = zi – zi –j
Where,
i = 1, 2, 3 ....... 2n + 1;
j = 1, 2, 3 ....... i
Step 3
Tabulate the approximations obtained in step-2 as shown in table below.
zi Qi, 0 Qi, 1 Qi, 2 Qi, 2n + 1
z0 Q0, 0
z1 Q1, 0 Q1, 1
Q2, 2
z2 Q2, 0 Q2, 1
z3 Q3, 0 Q3, 1 Q3, 2 Q2n+1, 2n+1
z2n Q2n, 0 Q2n+1, 2

Q2n+1, 1
z2n+1 Q2n+1, 0
Step 4
The Hermite interpolating polynomial is given as,
H(x) = Q0, 0 + Q1, 1 (x – x0) + Q2, 2 (x – x0)2 + Q3, 3 (x – x0)2 (x – x1) + Q4, 4 (x – x0)2 (x – x1)2
+ Q5, 5 (x – x0)2 (x – x1)2 (x – x2) - - - - - -
Q31. Use Hermite interpolation to construct an approximating polynomial for the following.
(a) x f(x) f'(x)
0.8 0.22363362 2.1691753
1.0 0.65809197 2.0466965
(b) Use the polynomial to approximate the value of ‘x’ in f(x) = sin(ex – 2) approximate f(0.9). Find the
absolute error.
Answer : Model Paper-2, Q10(b)
(a) Given,
x0 = 0.8, x1 = 1.0
f(x0) = 0.22363362, f(x1) = 0.65809197
f '(x0) = 2.1691753, f '(x1) = 2.0466965.
According to Hermite interpolation,
Qi,j – 1 – Qi – 1, j – 1
Qi,j = zi – zi – j ... (1)
z0 = z1 = x0 = 0.8
z2 = z3 = x1 = 1.0
Q0,0 = f(x0) = 0.22363362
Q1,0 = f(x0) = 0.22363362
Q2,0 = f(x1) = 0.65809197
Q3,0 = f(x1) = 0.65809197
Q1,1 = f '(x0) = 2.1691753
Q3,1 = f '(x1) = 2.0466965
From equation (1),
Q2,0 – Q1,0 0.65809197 – 0.22363362
Q2,1 = z2 – z1 = 1.0 – 0.8 = 2.17229175.
Q2,2–1 – Q2–1, 2–1 Q2,1 – Q1,1
Q2,2 = z2 – z2–2 = z – z = 2.17229175 – 2.1691753
1.0 – 0.8 = 0.01558225
2 0
Q3,1 – Q2,1 2.0466965 – 2.17229175

Q3,2 = z3 – z = 1.0 – 0.8 = – 0.62797625
Q3–2 – Q2–2 – 0.6279625 – 0.01558225

Q3,3 = z3 – z0 = 1.0 – 0.8 = – 3.2177925.
z f(z) First divided Second divided Third divided

difference difference difference
x0 = z0 = 0.8 Q0,0 = 0.22363362
Q1,1= 2.1691753
x0 = z1 = 0.8 Q1,0 = 0.22363362 Q2,2= 0.01558225
Q2,1=2.17229175 Q3,3= – 3.2177925
x1 = z2 = 1.0 Q2,0 = 0.65809197 Q3,2= – 0.62797625
Q3,1=2.0466965
x1 = z3 = 1.0 Q3,0 = 0.65809197
Table
According to Hermite’s interpolation formula,
H3(x) = Q0,0 + Q1,1 (x – x0) + Q2,2 (x – x0)2 + Q3,3 (x – x0)2 (x – x1) ... (2)
H3(x) = 0.22363362 + 2.1691753 (x – 0.8) + 0.01558225 (x – 0.8)2 – 3.2177925 (x – 0.8)2 (x – 1.0).
f(x) = sin(ex – 2)
f' (x) = cos(ex – 2) (ex)
⇒ f' (x) = ex cos(ex – 2)
f(x0) = f(0.8) = sin(e0.8 – 2) = 0.2236336
f(x1) = f(1.0) = sin(e1.0 – 2) = 0.65809196
f '(x0) = f ' (0.8) = e0.8 cos(e0.8 – 2) = 2.16917527
f '(x1) = f '(1.0) = e1.0 cos(e1.0 – 2) = 2.04669647
Hermite interpolating polynomial is,
H3(x) = 0.22363362 + 2.1691753 (x – 0.8) + 0.01558225 (x – 0.8)2 – 3.2177925 (x – 0.8)2 (x – 1.0).
Substituting x = 0.9 in the above equation,
H3(0.9) = 0.22363362 + 2.1691753 (0.9 – 0.8) + 0.01558225 (0.9 – 0.8)2 – 3.2177925 (0.9 – 0.8)2 (0.9 – 1.0).
\ H3(0.9) = 0.443924765
\ H3(0.9) is the approximate to f(x).
f(x) = sin(ex – 2)
⇒ f(0.9) = sin(e0.9 – 2)
= 0.44359243.
\ f(0.9) = 0.44359243 is the actual error value to f(x).
|f(0.9) – H3(0.9)| = |0.44359243 – 0.443924765|
= | – 3.32326229 × 10–4|
= 0.000332326
Q32. Use the following values and five digit rounding arithmetic to construct the Hermite interpolating
polynomial to approximate sin 0.34.

x sinx Dx(sin x) = cos x
0.30 0.29552 0.95534
0.32 0.31457 0.94924
0.35 0.34290 0.93937
Given,
x0 = 0.30, x1 = 0.32, x2 = 0.35
f(x0) = 0.29552, f(x1) = 0.31457, f(x3) = 0.34290
f '(x0) = 0.95534, f '(x1) = 0.94924, f '(x1) = 0.93937
Qi, j – 1 – Qi –1, j – 1
Qi,j = zi – zi – j ... (1)
z0 = z1 = x0 = 0.30
z2 = z3 = x1 = 0.32
z4 = z5 = x2 = 0.35
Q0,0 = f(x0) = 0.29552
Q1,0 = f(x0) = 0.29552
Q2,0 = f(x1) = 0.31457
Q3,0 = f(x1) = 0.31457
Q4,0 = f(x2) = 0.34290
Q5,0 = f(x2) = 0.34290
Q1,1 = f '(x0) = 0.95534
Q3,1 = f '(x1) = 0.94924
Q5,1 = f '(x2) = 0.93937.
From equation (1),

Q2,1 –1 – Q2–1, 1–1 Q2,0 – Q1,0
Q2,1 = z2 – z2 –1 = z – z = 0.31457 – 0.29552
0.32 – 0.30 = 0.9525
2 1
Q4,0 – Q3,0 0.34290 – 0.31457

Q4,1 = z4 – z3 = 0.35 –0.32 = 0.9443333.

Q2,1 – Q1,1
Q2,2 = z2 – z0 = 0.9525 – 0.95534
0.32 – 0.30 = – 0.142
Q3,1 – Q2,1
Q3,2 = z3 – z1 = 0.94924 – 0.9525
0.32 – 0.3 = – 0.163
Q4,1 – Q3,1
Q4,2 = z4 – z2 = 0.9443333 – 0.94924
0.35 – 0.32 = – 0.163556
Q5,1 – Q4,1
Q5,2 = z5 – z3 = 0.93937 – 0.9443333
0.35 – 0.32 = – 0.165443.
Q3,2 – Q2,2 – 0.163 – (– 0.142)
Q3,3 = z3 – z0 = 0.32 – 0.30 = – 1.05
Q4,2 – Q3,2 – 0.163556 – (– 0.163)
Q4,3 = z4 – z1 = 0.35 – 0.3 = – 0.01112
Q5,2 – Q4,2 – 0.165443 – (– 0.163556)
Q5,3 = z5 – z2 = 0.35 – 0.32 = – 0.0629
Q4,3 – Q3,3 – 0.01112 – (– 1.05)
Q4,4 = z4 – z0 = 0.35 – 0.3 = 20.776
Q5,3 – Q4,3 – 0.0629 – (–0.01112)
Q5,4 = z5 – z1 = 0.35 – 0.3 = – 1.0356
Q5,4 – Q4,4
Q5,5 = = –1.0356
z5 – z0
– 20.776
0.35 – 0.3 = – 436.232
z f(z) First divided Second divided Third divided Fourth divided Fifth divided
difference difference difference difference difference
x0 = z0 = 0.3 Q0,0 = 0.29552
Q1,1= 0.95534
x0 = z1 = 0.3 Q1,0 = 0.29552 Q2,2 = – 0.142
Q2,1 = 0.9525 Q3,3 = – 1.05
x1= z2 = 0.32 Q2,0 = 0.31457 Q3,2 = –0.163 Q4,4 =20.776
Q3,1= 0.94924 Q4,3= – 0.01112 Q5,5= –436.232
x1= z3 = 0.32 Q3,0= 0.31457 Q4,2 = –0.163556 Q5,4= –1.0356
Q4,1= 0.9443333 Q5,3 = –0.0629
x2= z4 = 0.35 Q4,0 = 0.34290 Q5,2 = – 0.165443
Q5,1 = 0.93937
x2= z5 = 0.35 Q5,0 = 0.34290
Table
H5(x) = Q0,0 + Q1,1 (x – x0) + Q2,2 (x – x0)2 + Q3,3 (x – x0)2 (x – x1)+ Q4,4 (x – x0)2 (x – x1)2
+ Q5,5 (x – x0)2 (x – x1)2 (x – x2).
Substituting the corresponding values in the above equation,
H5(x) = 0.29552 + 0.95534(x – 0.3) + ( – 0.142) (x – 0.3)2 + ( – 1.05) (x – 0.3)2 (x – 0.32)
+ 20.776 (x – 0.3)2 (x – 0.32)2 + (– 436.232) (x – 0.3)2 (x – 0.32)2 (x – 0.35)
Substituting, x = 0.34 in the above equation,
H5(0.34) = 0.29552 + 0.95534 (0.34 – 0.3) – 0.142(0.34 – 0.3)2 – 1.05 (0.34 – 0.3)2 (0.34 – 0.32)
+ 20.776(0.34 – 0.3)2 (0.34 – 0.32)2 – 436.232(0.34 – 0.3)2 (0.34 – 0.32)2 (0.34 – 0.35).
\ H5(0.34) = 0.33348888
Q33. Use Hermite interpolation to construct an approximating polynomial for the following data.
(a) x f(x) f’(x)
8.3 17.56492 3.116256
8.6 18.50515 3.151762
(b) Use the polynomial for the given value of ‘x’ to approximate f(x) = xlnx and calculate the absolute
error f(8.4) with the function.
Answer :
(a) Given,
x0 = 8.3, x1 = 8.6
f(x0) = 17.56492, f(x1) = 18.50515
f '(x0) = 3.116256, f '(x1) = 3.151762
Qi,j – 1 – Qi – 1, j – 1
Qi,j = zi – zi – j ... (1)
And,
z0 = z1 = x0 = 8.3
z2 = z3 = x1 = 8.6
Q0,0 = f(x0) = 17.56492
Q1,0 = f(x0) = 17.56492
Q2,0 = f(x1) = 18.50515
Q3,0 = f(x1) = 18.50515
Q1,1 = f '(x0) = 3.11625600
Q3,1 = f '(x1) = 3.151762
From equation (1),
Q2,1–1 – Q2–1, 1–1 Q2,0 – Q1,0
Q2,1 = z2 – z2–1 = z – z = 18.50515 – 17.56492
8.6 – 8.3 = 3.1341
2 1
Q2,2–1 – Q2–1, 2–1 Q2,1 – Q1,1

Q2,2 = z2 – z2–2 = z – z = 3.1341 – 3.116256
8.6 – 8.3 = 0.05948
2 0
Q3, 3–2 – Q3–1, 2–1 Q3, 1 – Q2,1

Q3,2 = z3 – z3–2 = z – z = 3.151762 – 3.1341
8.6 – 8.3 = 0.058873333
3 1
Q3, 3–1 – Q3–1, 3–1 Q3,2 – Q2,2

Q3,3 = z3 – z3–3 = z – z = 0.058873333 – 0.05948
8.6 – 8.3 = – 0.002022222.
3 0

z f(z) First divided Second divided Third divided
difference difference difference
x0 = z0 = 8.3 Q0,0 = 17.56492
Q1,1 = 3.116256
x0 = z1 = 8.3 Q1,0 = 17.56492 Q2,2 = 0.05948
Q2,1 = 3.1341 Q3,3 = – 0.002022222
x1 = z2 = 8.6 Q2,0 = 18.50515 Q3,2= 0.05887333
Q3,1 = 3.151762
x1 = z3 = 8.6 Q3,0 = 18.50515
Table
H3(x) = Q0,0 + Q1,1 (x – x0) + Q2,2 (x – x0)2 + Q3,3 (x – x0)2 (x – x1) ... (2)
H3(x) = 17.56492 + 3.116256(x – 8.3) + 0.05948(x – 8.3)2 – 0.002022222 (x – 8.3)2 (x – 8.6).
f(x) = x ln x
J N
f ' (x) = x KK 12 OO + ln (x)
L P
⇒ f ' (x) = x + ln (x)
Let,
x0 = 8.3, x1 = 8.6
f(x0) = 8.3 ln (8.3) = 17.56492
f(x1) = 8.6 ln (8.6) = 18.50515
f ' (x0) = 1 + ln (8.3) = 3116256
f '(x1) = 1 + ln (8.6) = 3.151762
Hermite interpolating polynomial corresponding to the above values is,
H3(x) = 17.56492 + 3.116256 (x – 8.3) + 0.05948 (x – 8.3)2 – 0.002022222 (x – 8.3)2 (x – 8.6) ... (3)
⇒ H3(8.4) = 17.56492 + 3.116256 (8.4 – 8.3) + 0.05948 (8.4 – 8.3)2 – 0.002022222 (8.4 – 8.3)2 (8.4 – 8.6).
\ H3(8.4) = 17.8771244.
Actual Value
f(x) = x ln x
f(8.4) = (8.4) ln (8.4)
⇒ f(8.4) = 17.87714633
Absolute error = |f(8.4) – H3(8.4)|
= | 17.87714633 – 17.8771244|
= 0.000022
2.5 Cubic Spline Interpolation

Q34. Write the steps involved in natural cubic spline.
Answer :
Natural Cubic Spline
Let, Sj(x) be the cubic polynomial defined on the sub-interval [xj, xj+1] for j = 0, 1, ...... n – 1.
Step 1
The natural cubic spline polynomial is defined as,
Sj(x) = aj + bj (x – xj) + cj(x – xj)2 + dj(x – xj)3, (xj ≤ x ≤ xj+1).
Step 2
The boundary conditions are,
s''(x0) = s''(xn) = 0.
Step 3
Ax = b
Where,
A is the (n + 1) × (n + 1) matrix and is given as,
1 0 0 0
h0 2(h0+h1) h1
0 h1 2(h1+ h2) h2 0
A =
hn–1
0 0 0 1
b is the vector given as,

RS 0 VW
SS WW
SS 3 3 W
SS hi + i (ai + 2 – ai + 1) – hi (ai + 1 – ai)WWW
b = SS h WW
SS WW
SS h WW
SS 0 WW
TR V X
SSc0WW
SS WW
c1
x = SSS WWW
SS h WW
SScnWW
T X
Where,
hi = xi+1 – xi
Step 4
The constants b0, d0 can be calculated as,
hj
bj = h1 (aj+1 – aj) – 3 (cj+1 + 2cj)
j
c j+1 – c j
dj = 3h
j
Step 5
Substitute the corresponding values in the natural cubic spline polynomial in order to obtain the required result.
Q35. Construct the clamped cubic spline for the following data and the fact f '(0.1) = – 2.8004996 and f'(0.3)
= – 2.9734038.
x f(x)
0.1 – 0.29004996
0.2 – 0.56079734
0.3 – 0.81401972
Given data is,

x0 = 0.1, x1 = 0.2, x2 = 0.3
f(0.1) = – 0.29004996
f(0.2) = – 0.56079734
f(0.3) = – 0.81401972.
f ' (0.1) = – 2.8004996
f ' (0.3) = – 2.9734038.
The given data consists of two cubic splines defined in the intervals [0.1, 0.2] and [0.2, 0.3].
Cubic spline in the interval [0.1, 0.2].
S0(x) = a0 + b0 (x – x0) + c0 (x – x0)2 + d0 (x – x0)3
⇒ S0(x) = a0 + b0 (x – 0.1) + c0 (x – 0.1)2 + d0 (x – 0.1)3 ... (1)
Cubic spline in the interval [0.2, 0.3]
S1(x) = a1 + b1 (x – x1) + c1 (x – x1)2 + d1 (x – x1)3
⇒ S1(x) = a1 + b1 (x – 0.2) + c1 (x – 0.2)2 + d1 (x – 0.2)3 ... (2)
Differentiating equation (1) with respect to x,
S'0(x) = b0 + 2c0 (x – 0.1) + 3d0 (x – 0.1)2 ... (3)
For x = 0.1
⇒ S'0(0.1) = b0 + 2c0 (0) + 3d0 (0)
\ f '0(0.1) = S'0 (0.1) = b0
\ b0 = – 2.8004996. ... (4)
Differentiating equation (2) with respect to x,
S'1(x) = b1 + 2c1(x – 0.2) + 3d1(x – 0.2)2 ... (5)
For x = 0.3
S'1 (0.3) = b1 + 2c1(0.3 – 0.2) + 3d1(0.3 – 0.2)2
⇒ S'1 (0.3) = f ' (0.3) = b1 + 0.2 c1 + 0.03 d1
\ b1 + 0.2 c1 + 0.03 d1 = – 2.9734038 ... (6)
Differentiating equations (3) and (5) with respect to x,
S0''(x) = 2c0 + 6d0 (x – 0.1)
S1''(x) = 2c1 + 6d1 (x – 0.2)
The cubic spline must agree with the data at nodes,
(i) S(0.1) = f(0.1)
a0 + 0 + 0 + 0 = – 0.29004996
\ a0 = – 0.29004996 ... (7)
(ii) S(0.2) = f(0.2)
⇒ S(0.2) = a0 + b0 (0.2 – 0.1) + c0 (0.2 – 0.1)2 + d0 (0.2 – 0.1)3
⇒ S0 (0.2) = a0 + 0.1 b0 + 0.01 c0 + 0.0001 d0
\ a0 + 0.1 b0 + 0.01 c0 + 0.0001 d0 = – 0.56079734 ... (8)
(iii) S1(0.2) = f(0.2)
⇒ S1(0.2) = a1 + b1 (0) + c1 (0)2 + d1 (0)3
⇒ S1 (0.2) = a1
\ a1 = – 0.56079734 ... (9)
(iv) S1(0.3) = f(0.3)
S1(0.3) = a1 + b1 (0.1) + c1 (0.1)2 + d1 (0.1)3
⇒ a1 + 0.1b1 + 0.01 c1 + 0.001 d1 = – 0.81401972 ... (10)
Clamped Conditions
(i) S0''(0.2) = S1''(0.2)
⇒ S0''(x) = 2c0 + 6d0 (x – 0.1)
⇒ S0''(0.2) = 2c0 + 6d0 (0.2 – 0.1)
\ S0''(0.2) = 2c0 + 6d0 (0.1)
And,
S1''(x) = 2c1 + 6d1 (x – 0.2)
S1''(0.2) = 2c1 + 6d1 (0.2 – 0.2)
S1''(0.2) = 2c1
\ 2c0 + 0.6 d0 = 2c1 ... (11)
(ii) S0'(0.2) = S1'(0.2)
S0'(0.2) = b0 + 2c0 (0.2 – 0.1) + 3d0 (0.2 – 0.1)2 [a From equation (3)]
⇒ S0'(0.2) = b0 + 2c0 (0.1) + 3d0 (0.1)2
S1'(0.2) = b1 + 2c1 (0.2 – 0.2) + 3d1 (0.2 – 0.2)2
⇒ S1'(0.2) = b1
\ b0 + 0.2 c0 + 0.03d0 = b1 ... (12)
Solving the equations,
Then,
a0 = – 0.29005
b0 = – 2.8005
c0 = 3.38169
d0 = – 24.51434
a1 = – 0.56080
b1 = – 2.19770
c1 = – 3.97261
d1 = 6.27414
S0(x) = – 0.29005 – 2.8005 (x – 0.1) + 3.38169 (x – 0.1)2 – 24.51434 (x – 0.1)3, E[0.1, 0.2]
S1(x) = – 0.56080 – 2.1977 (x – 0.2) – 3.97261 (x – 0.2)2 + 6.27414 (x – 0.2)3, E[0.2, 0.3].
Q36. (a) Construct the natural cubic spline for the following data.
x f(x)
– 0.5 – 0.0247500
– 0.25 0.3349375
0 1.1010000
J 1N J 1N
(b) Using the cubic spline constructed in the part a, approximate f KK– 3 OO and f' KK– 3 OO for f(x) = x3 +
4.001 x2 + 4.002 x + 1.101 and calculate the actual error. L P L P
Answer :
Given,
x0 = – 0.5, x1 = – 0.25, x2 = 0
f(x0) = – 0.0247500, f(x1) = 0.3349375, f(x2) = 1.1010000
Let, S(x) be a cubic polynomial and it is denoted by Sj(x).
According to natural cubic spline,
Sj(x) = aj + bj (x – xj) + cj (x – xj)2 + dj (x – xj)3 ... (1)
Where,
j = 0, 1, 2 ....... n – 1.
For j = 0,
S0(x) = a0 + b0 (x – x0) + c0(x – x0)2 + d0 (x – x0)3 ... (2)
For j = 1,
S1(x) = a1 + b1 (x – x1) + c1(x – x1)2 + d1(x – x1)3 ... (3)
hi = xi + 1 – xi , i = 0, 1, 2 ........ n – 1
⇒ h0 = x0+1 – x0

= x1 – x0
= – 0.25 – (– 0.5)
⇒ h0 = 0.25
And
h1 = x1 + 1 – x1
= x2 – x1
= 0 – (– 0.25)
⇒ h1 = 0.25
Ax = b
Where,
RS V
SS1 0 0 WW
A = SSh0 2 (h0 + h1) h1WWW
SS W
S0 0 1 WW 3 × 3
T X
RS V
SS 1 0 0 WW
⇒ A = SS0.25 2 (0.5) 0.25WWW
SS W
S 0 0 1 WW 3 × 3
T X
RS 0 VW
SS WW
SS 3 3 WW
b = SS h1 2 ( a – a 1) – ( a
h0 1 0 WW – a )
SS 0 WW
T X 2 ×1
RS VW
SS 0 WW
SS 3 3 WW
⇒ b = SS 0.25 ( 1 . 10100000 – 0 . 3349375 ) – 0.25 ( 0. 33493751 – 0 .0247500 ) WW
SS 0 WW
T X
RS VW
SS 0 W
⇒ b = SS9.19275 –4.31625WWW
SS WW
S 0 W
T X
RS VW
SS 0 WW
⇒ b = SS4.8765WW
SS W
S 0 WW
T X
RSc VW
SS 0WW
x = SSc1WW
SS WW
Sc2W
T X
Where c2 = 0
RS V R V RS V
SS 1 0 0 WW SSc0WW SS 0 WWW
W S W
⇒ SS0.25 1 0.25WW SSc1WW = SS4.8765WW
SS W S W SS W
S 0 0.25 0 WW3×3 SS0 WW 3×1 S 0 WW
T X T X T X
RS VW RS V
SS c0 + 0 + 0 WW SS 0 WWW
⇒ SS0.25 c0 + c1 + 0WW = SS4.8765WW
SS WW SS W
S 0 W S 0 WW
T X T X
c0 = 0.000000
0.25 c0 + c1 = 4.87650000
⇒ c1 = 4.87650000
Solving for b and d values.
h
b0 = h1 (a1 – a0) – 30 (c1 + 2c0)
0
= 0.125 (0.3349375 – (– 0.02475)) – 0.325 (4.8765 + 0)
= 0.125 (0.3596875) – 1.219125

3
⇒ b0 = 1.43875 – 0.406375
⇒ b0 = 1.03237500
h
b1 = h1 (a2 – a1) – 31 (c2 + 2c1)
1
= 0.125 (1.1010000 – 0.3349375) – 0.325 (2(4.8765))
= 3.06425000 – 0.81275000
⇒ b1 = 2.25150000
d0 = 31h (c1 – c0)

0
= 3 (01.25) (4.87650000 – 0)
\ d0 = 6.50200000
d1 = 31h (c2 – c1)

1
= 3 (01.25) (– 4.87650000)
\ d1 = – 6.50200000
S0(x) = – 0.0247500 + 1.03237500 (x – (– 0.5)) + 0.00000 + 6.50200000 (x – ( – 0.5))3.
\ S0(x) = – 0.0247500 + 1.03237500 (x + 0.5) + 6.50200000 (x + 0.5)3.
S1(x) = 0.3349375 + 2.25150000 (x – (– 0.25)) + 4.87650000 (x – (– 0.25)2 – 6.50200000 (x + 0.25)3
\ S1(x) = 0.3349375 + 2.25150000 ( – x + 0.5)) + 4.87650000 (x + 0.25)2 – 6.50200000 (x + 0.05)3
f(x) = x3 + 4.001 x2 + 4.002x + 1.101
J N J N3 JK –1 NO 2 JK –1 NO
⇒ f KK –31 OO = KK –31 OO + 4.001 K 3 O + 4.002 K 3 O + 1.101
L P L P L P L P
J N
⇒ f KK –31 OO = 0.1745185
L P
f '(x) = 3x2 + 8.002 x + 4.002
J N J N2 JK –1 NO
⇒ f ' KK –31 OO = 3 KK –31 OO + 8.002 K 3 O + 4.002
L P L P L P
J N
\ f ' KK –31 OO = 1.668
L P
The equation of natural cubic spline in part (a) is,
S0(x) = – 0.0247500 + 1.03237500 (x + 0.5) + 6.50200000 (x + 0.5)3.
J N J N J N3
S0 KK –31 OO = – 0.0247500 + 1.03237500 KK –31 + 0.5OO + 6.50200000 KK –31 + 0.5OO
L P L P L P
J N
S0 KK –31 OO = 0.17741435
L P
J N
\ S0 KK –31 OO = 0.17741435 is the approximate to f(x).
L P
J –1 N J –1 N
f KK 3 OO – S0 KK 3 OO = |0.1745185 – 0.17741435|
L P L P
= | – 2.895 × 10–3|
J –1 N J –1 N
⇒ f KK 3 OO – S0 KK 3 OO = 2.895 × 10–3.
L P L P
J N
\ f KK –31 OO = 0.17451852 is the actual f(x).
L P

Internal Assessment
Objective type
I. Multiple Choice Questions
1. In linear interpolation, the polynomial of degree at most n passes through _____ points. [ ]
(a) n (b) n+1
(c) n/2 (d) n(n + 1)
2. p0,1,2 = _____ [ ]
1 1
(a) x 2 – x 0 [(x – x0)p1,2 – (x – x2)p0,1] (b) x 2 – x 0 [(x – x0)p1,2 + (x – x2)p0,1]
1 1
(c) x 2 + x 0 [(x – x0)p1,2 – (x – x0)p0,1] (d) x 2 – x 0 [(x – x0)p1,2 + (x – x0)p0,1]
3. The Lagrange's error formula causes the errors to exceed the theoritical error estimate due to assumption of –––.
[ ]
(a) Four-digit arithmetic (b) Infinite-digit arithmetic
(c) Floating point arithmetic (d) All of the above
–1
4. Approximation of f (x) using iterated interpolation is called [ ]
(a) Iterated inverse interpolation (b) Lagranges interpolation
(c) Spline interpolation (d) All of these
5. The coefficients of Newton forward difference form of the interpolating polynomical are tabulated along the _____
in the table [ ]
(a) First row (b) First column
(c) Diagonal (d) None of the above
6. Newton’s backward difference formula is [ ]
_ ks i ∇k f(x ) _ ks i ∇k f(x )
n n
(a) pn(x) = f[xn] – / (–1) k

n
(b) pn(x) = f[xn] + / (–1) k
n
k=1 k=1
n
–s n
_ –s
(c) pn(x) = f[xn] – / (–1) _ k i ∇k f(xn) (d) pn(x) = f[xn] +
k
/ (–1) k
k i ∇k f(xn)
k=1 k=1
7. The degree of osculating polynomial is at most M is [ ]

n n
/ mi + n
(a) (b) n / m i + n
i=0 i=0
/ mi (n + 1)
(c) (d) None of the above
i=0
8. The Hermite polynomial of degree at most 2n + 1 is [ ]

n n
(a) H (x) = / f (x ) H (x) – / f ' (x ) H| (x)

2n+1 j n,j j n,j
j=0 j=0
n n
(b) H2n+1(x) = / f (x ) H j n,j (x) + / f ' (x ) H|

j n,j (x)
j=0 j=0
n n
(c)
H2n+1(x) = / f ' (x ) H j n,j (x) – / f (x ) H|
j n,j (x)
j=0 j=0
n n
(d)
H2n+1(x) = / f ' (x ) H j n,j (x) + / f (x ) H|
j n,j (x)
j=0 j=0

9. If f∈ C [a, b] with maxa ≤ x ≤ b |f (x) = M and S is the unique clamped cubic spline interpolant to f with respect to
4 iv
nodes x0 < x1 < - - - - - xn then, |f(x) – S(x)| = ––––––. [ ]

5M 5M min
(a) 384 (b) 384 0 # j # n – 1 (xj+1 – xj)
4
5M max
(c) 384 0 # j # n – 1 (xj+1 – xj)
4
(d) None of the above
10. From the properties of clamped spline interpolant ‘S’ for function ‘f’, defined at [a, b] then [ ]
h h
(a) f '(a) = h1 (a1 + a0) + 30 (2c0 + c1) (b) f '(a) = h1 (a1 – a0) + 30 (2c0 + c1)
0 0
h h
f '(a) = h1 (a1 + a0) – 30 (2c0 + c1)
(c) (d) f '(a) = h1 (a1 – a0) – 30 (2c0 + c1)
0 0
II. Fill in the Blanks

1. If f is defined and continuous on [a, b], for each e > 0, p(x) exists with _____.
2. The unique polynomial p(x) of degree atmost ‘n’ is given by _____.
3. The bound for the error involved in approximating a function by an interpolating poynomial is _____.
4. If p(x) is a polynomial of degree atmost ‘n’, then the (n + 1)st derivative of (p(n + 1) (x1) is identically_____.
5. The process of generating the interpolating polynomials recursively is called _____.
6. The divided differences of f with respect to x0, x1 - - - - xn are used to express pn(x) as _____
7. The zeroth divided difference of function f with respect to xi is _____.
8. If f(1.0) = 0.7651977 and f(1.3) = 0.6200860 then the f[x0, x1] = –––––.
3
9. The backward difference, for the sequence {p n} n = 0 is defined by _____.
10. _____ polynomials are generalized by both Taylor and the Lagrange polynomials.
KEY
I. Multiple Choice Questions
1. (b) 2. (a) 3. (b) 4. (a) 5. (c)
6. (d) 7. (a) 8. (b) 9. (c) 10. (d)
II. Fill in the Blanks

1. |f(x) – p(x)| < e, ∀ x in [a, b]
n
2. / f (xk) Ln,k (x)

k=0
f n + 1 (x (x))
3. (n + 1) ! (x – x0) (x – x1) ...... (x – xn)
4. zero
5. Neville’s method
6. pn(x) = a0 + a1(x – x0) + a2(x – x0) (x – x1) + ....... + an(x – x0) ....... (x – xn – 1)
7. f[xi] = f(xi)
8. –0.4837057
9. ∇pn = pn – pn–1
10. Osculating

III. Very Short Questions and Answers Q6. If f(x) = e , find the value of f(0.43).
2x
Q1. Write the first degree Lagrange interpolating Answer :

polynomial. Given,
Answer : f(x) = e2x
The first degree Lagrange interpolating polynomial ⇒ f(0.43) = e2(0.43) = 2.36316.
passing through (x0, y0) and (x1, y1) is given as,
Q7. What are the boundary conditions for natural
P(x) = L0(x) f(x0) + L1(x) f(x1)
cubic spline?
Where,
Answer :
x–x
L0(x) = x – x1
0 1 The boundary conditions for natural cubic spline are
x – x0 given as,
L1(x) = x – x
1 0
s''(x0) = s''(xn) = 0.
Q2. Find L0(x) which passes through the points
Q8. Write the formula for Cubic Spline interpolation.
(2, 4) and (5, 1).
Answer :
Answer :
Let, s(x) be a cubic polynomial denoted as sj(x) in the
Given,
interval [xj, xj+1].
(x0, y0) = (2, 4)
Where,
(x1, y1) = (5, 1)
j = 0, 1, - - - - n – 1.
x–5 –1
\ L0(x) = 2 – 5 = 3 (x – 5) Then the cubic spline interplation can be given as,
Q3. If x0 = 1.0, x1 = 1, 3, Q0,0 = 0.7651977, Q1,0 = sj+1 (xj+1) = sj (xj+1) = aj + bj(xj+1 – xj) + cj(xj+1 – xj)2
0.6200860 then find Q1,1(x). + dj(xj+1 – xj) .
3
Answer :
(x – x0) Q1,0 – (x – x1) Q0,0 Q9. What is the value of ‘b’ in the vector equation
Q1,1(x) = x1 – x0
used in the Natural Spline?
(x – 1.0) (0.6200860) – (x – 1.3) (0.7651977)
= 1.3 – 1.0 Answer :
– 0.1451117x + 0.37467101 The vector equation is given as,

\ Q1,1(x) = 0.3 Ax = b
16 Where,
Q4. If P3(x) = 1 + 4x + 4x(x – 0.25) + 3 x(x – 0.25)
RS VW
(x – 0.5) then find the value of P3(x) at 0.75. SS 0 WW
SS 3 3 WW
Answer : SS ( a
h1 2 1– a ) – (
h0 1 0a – a ) WW
S WW
Given, b = SS h WW
SS 3 3 W
16 SS (a – a ) – h (an–1 – an–2)WW
P3(x) = 1 + 4x + 4x(x – 0.25) + 3 x(x – 0.25) (x – 0.5) SS hn–1 n n–1 n –2 WW
S 0 W
⇒ P3(0.75) = 1 + 4(0.75) + 4(0.75) (0.75 – 0.25) T X
Q10. If x0 = 1, x1 = 2, then find the values of S0(x), S1(x).
16
+ 3 (0.75) (0.75 – 0.25) (0.75 – 0.5)
Answer :
\ P3 (0.75) = 6. S0(x) = a0 + b0(x – x0) + c0(x – x0)2 + d0(x – x0)3
Q5. Write the Hermite polynomial of degree at most ⇒ S0(1) = a0 + b0(x – 1) + c0(x – 1)2 + d0(x – 1)3
(2n + 1).
And
Answer :
Let f ∈ [a, b] and x0, - - - - -, xn ∈ [a, b] are distinct. Then S1(x) = a1 + b1(x – x1) + c1(x – x1)2 + d1(x – x1)3
the Hermite polynomial of degree atmost (2n + 1) is given as,
n n
⇒ S1(1) = a1 + b1(x – 2) + c1(x – 2)2 + d1(x – 2)3
H2n+1(x) = / f (x ) Hj n, j (x) + / f ' (x ) H|
j n, j (x)
j=0 j=0

Maths NA 2

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Maths NA 2

Uploaded by

Copyright:

Available Formats

2

LEARNiNG OBJEC TiVES

v Interpolation and Lagrange Polynomial

v Data Approximation and Neville's Method

v Cubic Spline Interpolation.

INT RODUC TiON

Given function is,

f(x1) = f(0.5) = e0.5 = 1.6487212

\ f(0.25) = e0.25 = 1.28402.

Given function is,

f(x0) = f(0.5) = e0.5 = 1.64872

f(x1) = f(1) = e1 = 2.71828.

⇒ P1(x) = 2.13912 x + 0.57916

\ f(0.75) = e0.75 = 2.117.

Given function is,

f(x1) = f(1) = e1 = 2.718281.

f(x2) = f(2) = e2 = 7.38905.

= (x2 – 3x + 2) (0.5) – 2.71828 (x2 – 2x) + 3.69452 (x2 – x)

= x2 (0.5 – 2.71828 + 3.69452) + x(– 1.5 + 5.43656 – 3.69452) + 1

\ f(0.25) = e0.25 = 1.28402

f(0.75) = e0.75 = 2.117.

\ Q1,1 (1.5) = 0.8521

(1.5 – 0) 3.5 – (1.5 – 2) (2.5)

(1.5 – 0) P1,2,3 (1.5) – (1.5 – 3) P0, 1,2 (1.5)

Answer : Model Paper-2, Q4

x0 = – 0.1, x1 = 0.0, x2 = 0.2, x3 = 0.3

f(x0) = 5.30000, f(x1) = 2.00000, f(x2) = 3.19000, f(x3) = 1.00000

The first divided difference corresponding to x0 and x1 is given as,

pn(x) = f [x0] + / f [x , x ...... x ] [(x – x ) ...... (x – x

= 0.125 [ – 0.02475000 – (– 0.07181250)]

= 0.125 [1.43875 – 1.18825]

= 0.125 [6.502 – 1.002]

Substituting the corresponding values in equation (3),

x0 = – 0.75, x1 = – 0.5, x2 = – 0.25, x3 = – 0, x4 = –31

f(x0) = – 0.07181250, f(x1) = – 0.02475000, f(x2) = 0.33493750, f(x3) = 1.10100000.

x f(x) ∇f ∇2f ∇3f

Substituting the corresponding values in equation (1),

J N KKJ – 4 NOO JKK – 4 + 1NOO

SIA PUBLISHERS and DISTRIBUTORS PVT. LTD. 95

H(x) = Q0,0 + Q1,1 (x – x0) + Q2,2 (x – x0)2 + Q3,3 (x – x0)2 (x – x1)

96 SIA PUBLISHERS and DISTRIBUTORS PVT. LTD.

f'(x1) = 3(1.05) e(1.05) + 3e(1.05) – 2e2(1.05)

According to Hermite's interpolation formula,

= 01.3 (18.50515 – 17.56492) – 03.3 (2(0))

100 SIA PUBLISHERS and DISTRIBUTORS PVT. LTD.

2.1 Interpolation and the Lagrange Polynomial

P(x) = f(x0) Ln,0(x) + f(x1) Ln,1(x) + ......... + f(xn) Ln,n(x)

Given function is,

f(x1) = 1 + 0.6 = 1.264911

f(x2) = 1 + 0.9 = 1.378404

The interpolating polynomial of degree at most one is given as,

P1(x) = L0(x) f(x0) + L1(x) f(x1) ... (1)

SIA PUBLISHERS and DISTRIBUTORS PVT. LTD. 101

f(0.45) = 1 + 0.45 = 1.204153

Absolute error is,

|f(0.45) – P1(0.45)| = |1.204153 – 1.1986831| = 0.0054699.

The interpolating polynomial of degree atmost two is given as,

P2(x) = L0(x) f(x0) + L1(x) f(x1) + L2(x) f(x2) ... (2)

P1(x) = L0(x) f(x0) + L1(x) f(x1) ... (1)

Substituting the corresponding values in equation (1),

f(0.45) = tan(0.45) = 0.4830551

P1(0.45) = 1.1402280(0.45) = 0.5131026

Absolute error = | f(0.45) – P1(0.45)|