Вы находитесь на странице: 1из 4

MATH-501: Homework # 1

Due on Wednesday, February 11, 2015

Saket Choudhary

2170058637

1

Saket Choudhary

MATH-501 : Homework # 1

Contents

 

Problem # 1

 

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

3

1a

3

1b

3

1c

3

 

2 .

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

4

3 .

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

4

Saket Choudhary

MATH-501 : Homework # 1

Problem # 1

1a

1

sinx = p 0 + p 1 x Consider sin(x) p 1 x p + 0 | 2 = 1 (sin(x) p 1 x p 0 ) 2 dx In order to find, p 1 , p 0 we consider partial derivatives

and

1

Using Liebnitz’s formula in 1: 1 2(x)(sin(x) p 1 x p 0 )dx = 0

=

1

1 x.sin(x) p 1 x 2 p 0 xdx =

= 0 Thus, p 1 = 3(sin(1) cos(1))

Similarly using Leibnitz’s ruke on 2: 1 2(1)(sin(x) p 1 x p 0 )dx = 0 =p 0 = 0 (The first two terms

are odd terms and hence integrate to 0) p 0 is also justified since sin(x = 0) = 0 Hence sin(x) = 3(sin(1) cos(1))x

1

dp 1

d

1

1 (sin(x) p 1 x p 0 ) 2 dx

dp 1 d 1 − 1 ( sin ( x ) − p 1 x −

= 0

(1)

dp 0

d

1

1 (sin(x) p 1 x p 0 ) 2 dx = 0

dp 0 d 1 − 1 ( sin ( x ) − p 1 x −

(2)

0 =⇒ −x.cos(x) |

1 1 +

1

1 cos(x)dx 2p1

3

0 = ⇒ − x.cos ( x ) | 1 − 1 + 1 − 1

1b

Taylor approximation(degree 3) around t

 

0:

p 2 (t)

 

sin(0) + cos(0)

1!

(x 0) 1

+

sin(0)(x0) 2

+

=

=

2!

cos(0)(x0) 3

3!

+ R 4 + R 4 (t) where R 4 is o(t 4 ) remainer term.

p 2 (t) = t t 3

3!

1c

, sin(1) = sin(1) and sin( 1 ) = sin(

Given f (t) at t = 1, 1

3

3 , 1 we fit a degree3 polynomial for sin(x) using Legendre Polynomials.

3

1 3 )

1

Now, l 0 (x) = (x+ 1 3 )(x

( 2

3

3 1 )(x1)

)( 4 )(2)

3

= 9

16

1

(x + 3 )(x

3 1 )(x 1)

l 1 (x) = (x+1)(x− 1 3 )(x−1) 2 ( 3 )( −2 )( −4
l 1 (x) = (x+1)(x− 1 3 )(x−1)
2
(
3 )( −2
)( −4 )
3
3
1
l 2 (x) = (x+1)(x+ 3 )(x−1)
( 4 )( 2 3 ))( −2 )
3
3
1
l 3 (x) = (x+1)(x+ 3 )(x− 1 3 )
2
(2)( 4 )( 3 ))
3

27

= (x + 1)(x

16

3 1 )(x 1)

= (x + 1)(x + 3 )(x 1)

16

27

1

9

= 16 (x + 1)(x + 3 )(x 1 3 )

1

3

sin(x) = i=0 t i xl i (x)

1

Thus sin(x) = (sin(1))l 0 (x) + (sin( 3 ))l 1 (x) +

1

sin( 3 )l 2 (x) + sin(1)l 3 (x)

Saket Choudhary

MATH-501 : Homework # 1

(continued)

2

Given: u 1 = 1, u 2 = x, u 3 = x 2 w
Given:
u 1 = 1, u 2 = x, u 3 = x 2 w i (x) =
v i
where v i is given by:
v 1 = u 1 = 1
v
i
u 2 − u 2 ,v 1 v 1
v 3 = u 3 − u 3 ,v 1 v 1
u 3 ,v 2 v 2
v 2 =
v 1 ,v 1
v 1 ,v 1
v 2 ,v 2
1
x 2 (x− 1 2 )dx
1
1
v 3 = x 2 − (
1 x 2 dx + (x −
0
= x 2 − x +
1
0
2 )
0
1
(x− 2 ) 2 dx
6
Similarly, w 1 = 1
w 2 =
1
x−
2
1
12
1
w 3 = x 2 −x+
6
1
180
2
3 +
4
Pf =
(x − 1 2 ) + −4 (x 2 − x +
5
7
1
6 )

v 2 =

x

xdx

dx = x 1

1 2

2

1


0

1


0

v 2 = x − xdx d x = x − 1 1 2 2 1

Pf

3

= i=1 f, w i w i = (

1

1 xdx)1 + (

0

0

x(x 1 2 )dx)(x 1 2 ) + (

0

1

x(x 2 x + 6 )dx)(x 2 x + 1 6 )

1

3

By Weierstrass’ approxmiation theorem for > 0 there exists a polynomial p(x) such that p f = max|f (x) p(x)| < while a x b |E n (f)| = |E n (f) E n (p)|

= |

1 f (x)dx

0

i=1 n w i f(x i )|

= |

1 f (x)dx

0

1 p(x)dx + n i=1 w i p(x i ) i=1 n w i f(x i )|

0

 

1

n

(f (x) p(x))dx + i=1

w i (p(x i ) f(x i ))|

= |

0

Thus,

0

1

n

|E n (f)| = |

(f (x) p(x))dx + i=1

Applying triangular inequality,

w i (p(x i ) f(x i ))| (Since, E n (P)

= 0

E n (f) =

0

1

n

|(f (x) p(x))|dx + i=1

w i |(p(x i ) f(x i ))| ≈ p f

+ p f

 

Thus,

p f

2 and hence there exists a N > 0 such that |E n (f )| < when n > N