Problem 1 Decide if each of the following claims is true or false. Justify your answer with a brief argument (for example a reference to the textbook, including exercises done in class, a short proof, or a counterexample
(i) CEO, 1] is a Banach space when equipped with the norm 111112 = (fol If (x)I2 dx) .
(ii) co, is a Banach space when equipped with the norm 11116 = suNcom If (x)I.
(iii) If M is a closed subspace of a Hilbert space H and if M1 = {O}, then M = H.
(iv) If H is a Hilbert space with a complete orthonormal set (ca):°=1, then at, le„ defines an element in H.
(v) If f E al has Fourier coefficients c„(f),n EZ, then E°°,2__. len(f)12 < 00.
(vi) Every linear operator T: E F between two normed spaces E and F is bounded.
(vii) Every continuous linear operator T: E —> F between two normed spaces E and F is bounded.
Problem 2 Consider the vector space C[0,1] equipped with the inner product (Ls) = f (x)9(x) dr, f,s E C[0,Jol
and associated norm 11111 = (1, f)'12. Let M be the subspace of C[0,11 spanned by the functions th(x) = 1 and tz2(x) = x. Let h E C[0,1] be given by h(x) = f.
= I (x)9(x)ax, 1,9 e t.4u,
and associated norm 11111 = (f, f)'/2. Let M be the subspace of C[0,1] spanned by the functions ui(x) = 1 and u2(x) = x. Let h E C[0,1] be given by h(x) = f.
(1) Find an orthonormal basis for M.
(ii) Find a, b E C which minimize the norm IIh — au1 — bull'. Sketch the graphs of h and au' + bu2 in the same coordinate system.
(iii)Determine IIhlI, Ilh — aui — 611211, and Haul + WA, where a, b are as in question (ii). Problem 3 (16 points). Consider the Hilbert space L2(--7r, 7r) with its usual inner product (f, 9) = f (x)g(x)dx, f, g E L2(—/r,
Let (e„)°?__03 be the orthonormal basis for L2(—ir , Tr) consisting of the trigonometric functions
ei,(x) = 2T el", x EV
and let c„(f) = (f,e„) denote the Fourier coefficients of f E
Consider the functions f,g: C given by f(x) = IxI and g(x) = sin(x). You
may use that sin(x) = (e — e-lx)/2i, and you are not asked to prove this fact.
(1) Calculate 111112 and 119112.
(ii) Determine the Fourier coefficients c„(f) and c„(g) for all n E Z.
(iii) Determine act_cocn(f)cn(g).
Problem 4 (8 points). Consider the Hilbert space L2(-1,1) and the linear functional F on L2(-1, 1) given by1
F(f) = jilt f (t) dt, f E
Show that F is bounded and determine IIFII• Find f E L2(-1,1) such that Ilf II = 1 and IF(f)I=
Problem 5 Consider the Hilbert space e2 and consider the linear operator .9: ea e given by
S(Xj, X2) X3, X4) X5, X61 • • • ) = (X2) XII X4) X3) XIS) X5> • • • )) €2,
1Sx = y . (yn),00 1, y. . xn+i, n odd
x„_1, n even.
You are given, and you need not prove, the facts that S is linear and that Sx E €2 for all x E /2.
(i) Show that S E 2(12) and determine 11Sii•
(ii) Show that Ss = S and that S2 = I.
(iii) Put P = i(I + 5). Show that P is a projection, i.e., show that P = P and that P = P2. [Hint: You may use, without giving a proof, that I is Hermitian, i.e., I' = I. You may also use that the identity operator I acts as a multiplicative unit on 2(9), i.e., IT = TI = T for all T E sew)"
(iv) Put x= (1,2,3,4,3,2,1,0,0,0, ) E e.
Determine the vectors Px and x — Px and calculate the norms 114 IIPxII, and Ilx — Prd.
Problem 6 (10 points). Let 4 be the subspace of C2 consisting of all sequences x (x.):°_,1 which eventually are zero, ef. Example 2.7 in the textbook. Equip C2 and 4 with the norm Hall = (x, x)1/2 arising from the usual inner product on e.
Consider the linear functionals F,G: 4 —) C given by
2011VW= E x,„ G(x) = E x., x = (xa)wl E 4.
.m1 nalShow that G is bounded and that F is unbounded. Determine 11011.A comment about Fourier coefficients: Fourier coefficients for a function f E L2(—sr,rr) are at the lectures defined to be c„(f) = (f , en), where e„(x) = (21r)-112einz, x E I-7T, srbwhence f en(nen. The textbook uses Fourier coefficients denoted en, whichsatisfy f(x) Ec° cnenz. The two notions of Fourier coefficients are related via theformula: c„ = (2x)-1/2e„(f). In this exam, Fourier coefficients will always be of the first kind, that is calf) = (f,e„).
Attachment:- Find an orthonormal basis for M.zip