pdf

School

The Chinese University of Hong Kong *

*We aren’t endorsed by this school

Course

3230

Subject

Mathematics

Date

Nov 24, 2024

Type

pdf

Pages

4

Uploaded by MinisterGorilla8297

Report
MATH3230 Numerical Analysis Tutorial 1 1 Recall: The theorems and tools below will be frequently used in this course. Theorem 1 (Polynomials Theorem) . A polynomial of degree n has exactly n roots, and a polynomial of degree n which vanishes at ( n + 1) points is identically zero. Theorem 2 (Rolle’s theorem) . Let f ( x ) be continuous on [ a, b ] and differentiable in ( a, b ) , and f ( a ) = f ( b ) , then there exists ζ ( a, b ) such that f ( ζ ) = 0 . Theorem 3 (Mean-value theorem) . Let f ( x ) be continuous on [ a, b ] and its derivative f ( x ) be continuous in ( a, b ) , then there exists ζ ( a, b ) such that f ( b ) = f ( a ) + ( b a ) f ( ζ ) , i.e., f ( ζ ) = f ( b ) f ( a ) b a . Theorem 4 (Intermediate value theorem) . Let f ( x ) be continuous on [ a, b ] , then for any real number g that lies between f ( a ) and f ( b ) there exists ζ [ a, b ] such that f ( ζ ) = g . Theorem 5 (Taylor’s theorem in one dimension) . Let f ( x ) be continuously differentiable on [ a, b ] up to order n 1 and its n th derivative f ( n ) ( x ) be continuous in ( a, b ) . Then for any two points x 0 , x ( a, b ) , f ( x ) = f ( x 0 ) + f ( x 0 )( x x 0 ) + f ′′ ( x 0 ) 2! ( x x 0 ) 2 + · · · + f ( n ) ( x 0 ) n ! ( x x 0 ) n + R n ( x ) where R n ( x ) is the remainder of Taylor’s series and can be given by R n ( x ) = 1 n ! Z x x 0 f ( n +1) ( t )( x t ) n dt or R n ( x ) = f ( n +1) ( ζ ) ( n + 1)! ( x x 0 ) n +1 for some ζ lying between x 0 and x . Theorem 6 (Mean-value theorem for integrals) . Let f ( x ) and g ( x ) be continuous on [ a, b ] , and f ( x ) does not change signs in [ a, b ] . Then there exists ζ ( a, b ) such that Z b a f ( x ) g ( x ) dx = g ( ζ ) Z b a f ( x ) dx Theorem 7 (Taylor’s theorem in higher dimensions) . Let f ( x ) : R m R be sufficiently smooth in R m . Then for any two points x 0 , x , we have f ( x ) = f ( x 0 ) + Df ( x 0 ) T ( x x 0 ) + 1 2! ( x x 0 ) T D 2 f ( x 0 )( x x 0 ) + · · · where Df ( x ) is the gradient of f ( x ) and D 2 f ( x ) is the Hessian matrix of f ( x ) . Theorem 8 (Fundamental theorem of calculus) . Let F be differentiable in an open set R n and x . Then for all x such that the line segment connecting x and x lines inside , we have F ( x ) F ( x ) = Z 1 0 F ( x + t ( x x ))( x x ) dt . 1
2 Exercises: Please do the star problem (*) in tutorial class first and finish the rest after class. 1.* Let p, q : R R be two polynomials of degree n . Suppose p and q coincide on a distinct set of n + 1 real numbers { x i } n i =0 , i.e. p ( x i ) = q ( x i ) for all i = 0 , 1 , 2 , · · · , n. Using the polynomials theorem, show that p ( x ) = q ( x ) for all x R . Proof. Let f = p q , then f is a polynomial of at most degree n . Then we have f ( x i ) = 0 for all i = 0 , 1 , 2 ..., n . From polynomials theorem, we know this polynomial f of degree n that vanishes at n + 1 points is a constant function of zero, so f = 0 and p = q . 2. (a)* Determine whether we are allowed to use Rolle’s Theorem to guarantee the existence of some c in ( a, b ) such that f ( c ) = 0 . If not, explain why not. i. f ( x ) = cos( x ) , for x [0 , 2 π ] ; ii. f ( x ) = | x 2 | , for x [0 , 4] ; iii. f ( x ) = 1 /x 2 , for x [ 1 , 1] . (b) For f C n ( R ) and f ( x 0 ) = f ( x 1 ) = · · · = f ( x n ) = 0 for x 0 < x 1 < · · · < x n , show that there exists f n ( ζ ) = 0 for some ζ ( x 0 , x n ) . Solution. (a) We recall the condition of applying Rolle’s theorem is f ( x ) must be continuous in [ a, b ] , differ- entiable in ( a, b ) , and f ( a ) = f ( b ) . i. Since f ( x ) is continuous in [0 , 2 π ] , differentiable in (0 , 2 π ) , and f (0) = f (2 π ) = 1 , there exists c (0 , 2 π ) such that f ( c ) = 0 . ii. Since f ( x ) is not differentiable at x = 2 , Rolle’s theorem does not apply. iii. Since f ( x ) is undefined at x = 0 , Rolle’s theorem does not apply. (b) By Rolle’s theorem, there exist x 1 j ( x j , x j +1 ) , for j = 0 , · · · , n 1 such that f ( x 1 j ) = 0 for j = 0 , · · · , n 1 , and x 1 0 < x 1 1 < · · · < x 1 n 1 . Then by using Rolle’s Theorem sequentially on f k ( x ) for k = 1 , · · · , n 1 , we finally have there exists ζ ( x 0 , x n ) such that f n ( ζ ) = 0 . 3. (a)* Prove the mean value theorem using Rolle’s Theorem. (b) Let f, g : [ a, b ] R be continuous on [ a, b ] and differentiable on ( a, b ) . Assume g ( x ) > 0 for all x ( a, b ) . Using the mean value theorem, show that there exists c ( a, b ) such that f ( c ) g ( c ) = f ( b ) f ( a ) g ( b ) g ( a ) . Solution. (a) Let f ( x ) be continuous on [ a, b ] and f ( x ) be continuous in ( a, b ) , consider F ( x ) := f ( x ) f ( a ) f ( b ) f ( a ) b a ( x a ) . Then we have F ( a ) = F ( b ) = 0 . By Rolle’s theorem, there exists ζ ( a, b ) such that F ( ζ ) = f ( ζ ) f ( b ) f ( a ) b a = 0 , which implies f ( b ) = f ( a ) + ( b a ) f ( ζ ) . 2
(b) Consider G ( x ) := g ( x )( f ( b ) f ( a )) f ( x )( g ( b ) g ( a )) , we have G ( a ) = G ( b ) . Using the mean value theorem, there exists c ( a, b ) such that G ( c ) = 0 . This leads to g ( c )( f ( b ) f ( a )) f ( c )( g ( b ) g ( a )) = 0 . Since g ( x ) > 0 in ( a, b ) , we have g ( b ) g ( a ) > 0 , g ( c ) > 0 . Therefore we can derive from the equation above that f ( c ) g ( c ) = f ( b ) f ( a ) g ( b ) g ( a ) . 4. Using the intermediate value theorem to show the following. (a)* Show that the equation cos( x ) = 2 x has a solution in the interval [0 , π/ 2] . (b) Let f : [0 , 1] R be continuous such that f (0) = f (1) . Show that there exists c [0 , 1 / 2] such that f ( c ) = f ( c + 1 / 2) . Solution. (a) Let f ( x ) = cos( x ) 2 x , since f ( x ) is continuous and f (0) = 1 > 0 , f ( π/ 2) = 0 π < 0 , there exists ζ (0 , π/ 2) such that f ( ζ ) = 0 . Hence, there is a solution ζ to cos( x ) = 2 x in [0 , π/ 2] . (b) Consider g ( x ) = f ( x ) f ( x + 1 / 2) , since g ( x ) is continuous and g (0) = f (0) f (1 / 2) , g (1 / 2) = f (1 / 2) f (1) = g (0) , there exists c [0 , 1 / 2] such that g ( c ) = f ( c ) f ( c + 1 / 2) = 0 , i.e., f ( c ) = f ( c + 1 / 2) . In particular, c = 0 or 1 / 2 when f (0) = f (1 / 2) = f (1) . 5. (a)* Let F ( x, y ) = ( xy ) 2 + sin( x + y ) + 1 , given F (0 , 0) = 1 , using the fundamental theorem of calculus to compute F (1 , 1) by evaluating a line integral. Please specify all your steps clearly. (b) Using Taylor’s theorem to prove 1 + x < e x is valid for all real numbers x except 0 . Solution. (a) By definition, we have ∂F ( x, y ) ∂x = 2 xy 2 + cos( x + y ) , ∂F ( x, y ) ∂y = 2 x 2 y + cos( x + y ) . Then by the fundamental theorem of calculus, we have F (1 , 1) = F (0 , 0) + Z 1 0 2 t 3 + cos(2 t ) 2 t 3 + cos(2 t ) T 1 1 dt = 1 + Z 1 0 (4 t 3 + 2 cos(2 t )) dt = 2 + sin(2) . (b) Let f ( x ) = e x (1 + x ) , by the Taylor’s theorem, we have f ( x ) = f (0) + xf (0) + f 2 ( ζ ) 2 x 2 , for some ζ (0 , x ) when x > 0 and ζ ( x, 0) when x < 0 . Then as f (0) = 0 , f ( x ) = e x 1 where f (0) = 0 , and f ′′ ( x ) = e x > 0 for all x , we have f ( x ) = f 2 ( ζ ) 2 x 2 that is larger than 0 for all x except 0 which further implies 1 + x < e x for all x except 0 . 6. Let R n ( x ) = 1 n ! R x x 0 f ( n +1) ( t )( x t ) n dt . 3
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
  • Access to all documents
  • Unlimited textbook solutions
  • 24/7 expert homework help
(a)* Using the mean value theorem for integrals, show R n ( x ) can also be written as R n ( x ) = f ( n +1) ( ζ ) ( n + 1)! ( x x 0 ) n +1 . (b) Using integration by parts, show R n = 1 n ! f ( n ) ( x 0 )( x x 0 ) n + R n 1 . (c) Applying the integration by parts repeatedly, show f ( x ) = f ( x 0 ) + n X k =1 1 k ! f ( k ) ( x 0 )( x x 0 ) k + R n ( x ) . Solution. (a) By the mean value theorem for integrals there exists ζ ( x 0 , x ) such that R n ( x ) = f n +1 ( ζ ) n ! Z x x 0 ( x t ) n dt = f n +1 ( ζ ) ( n + 1)! ( x t ) n +1 x x 0 = f ( n +1) ( ζ ) ( n + 1)! ( x x 0 ) n +1 . (b) Recall the formula for integration by parts Z udv = uv Z vdu , and we apply it to the integral R n with u = ( x t ) n n ! , v = f ( n ) ( t ) , then we obtain R n ( x ) = 1 n ! h f ( n ) ( t )( x t ) n i t = x t = x 0 + Z x x 0 f ( n ) ( t ) ( x t ) n 1 ( n 1)! dt = 1 n ! f ( n ) ( x 0 )( x x 0 ) n + R n 1 ( x ) . (c) If the integrating parts is applied repeatedly on R j for j = n 1 , n 2 , · · · , 1 , we eventually obtain R n ( x ) = n X n =1 1 k ! f ( k ) ( x 0 )( x x 0 ) k + R 0 ( x ) . Since R 0 ( x ) = Z x x 0 f ( t ) dt = f ( x ) f ( x 0 ) , we have f ( x ) = f ( x 0 ) + n X k =1 1 k ! f ( k ) ( x 0 )( x x 0 ) k + R n ( x ) . 4