MATH 632. Functional Analysis (Fall 2006)

Meeting times: MWF, 9:00am-9:50am (MTH 0407)

Instructor: Professor Jonathan Rosenberg. His office is room 2114 of the Math Building, phone extension 55166, or you can contact him by email. His office hours are M and F 1-2, or by appointment.

Text: Peter Lax, Functional Analysis (Wiley). The book is available at a substantial discount at amazon.com and possibly other online bookstores, so shop around.

Prerequisite: MATH 631 (real analysis).

Catalog description:

Introduction to functional analysis and operator theory: normed linear spaces, basic principles of functional analysis, bounded linear operators on Hilbert spaces, spectral theory of selfadjoint operators, applications to differential and integral equations, additional topics as time permits.


Course Description:

Lax's book has a lot more in it than can be covered in one semester, but it's readable and gives you a very good feeling for what the subject is good for. I am planning to concentrate on the following topics:

If you can, try to look at the "applications" chapters, even if we don't have time to cover them in class. They are some of the best parts of the book, and show you the power of the subject.

Course Requirements and Grading Policy:

Homework will be assigned, collected, and graded, usually once a week. I will probably give a take-home final exam. Grades will have the following meaning:

It is your responsibility to turn the homework in on time.

Course Evaluation

We are asking for your help with on-line course evaluation. Please provide feedback on the course and the professor between Tue., Nov. 21 (12:00 am) and Wed., Dec. 13. Evaluations are anonymous and will not be available to faculty until next semester, so they cannot possibly affect your grade.

Tentative Schedule

(Some details to be filled in later if necessary)


Week Material Covered (Reading Assignment) Homework and Special Notes
8/30, 9/1.
(no class 8/28)
Ch. 1, 2.1 (read for review); Ch. 3-4 due 9/6:
  1. Use the Hahn-Banach Theorem to show that if V is a normed linear space over R (see 5.1 if you forget the definition) and if x is non-zero in V, then there is a linear functional f on V, of norm 1, for which f(x) is equal to the norm of x (and in particular is strictly positive). Deduce that linear functionals of norm 1 separate points, i.e., that if x and y are distinct points of V, then there is a linear functional f on V, of norm 1, for which f(x) is unequal to f(y).
  2. Suppose V is a vector space over R with a "positive cone"; that is, there is a subset V+ of V, closed under addition and multiplication by non-negative scalars, such that V = V+ - V+ and V+ and -V+ intersect only in {0}. A linear functional f on V is called positive if f is non-negative on V+. (Yes, I know; "non-negative" would be more descriptive, but this is the traditional terminology.) We write x y if y - x lies in V+. Suppose V has an order unit, that is, there is an element 1 in V such that for any x in V, there is a real number t > 0 such that tx 1. Show there is a positive linear functional f on V, with f(1) = 1. Next (this is harder), suppose x is a non-zero interior point in V+. Show there is a positive linear functional f on V, with f(1) = 1 and for which f(x) is strictly positive.
  3. Here's an application of the Hahn-Banach Theorem to economics, due basically to Debreu. Consider the vector space V = RN, whose dimension N is the number of goods in the economy, finite but large, and whose jth coordinate represents the amount of the jth good. Give this the positive cone V+ (in the sense of the exercise above) for which the coefficients of all goods are non-negative. We require that the amount of consumption x is equal to the amount of production y, and both lie in V+. (I.e., supply and demand are balanced, and no good is produced in negative amounts.) Assume that the production vector y must lie in a compact convex set Y in V+ (constraining what combinations of goods can be produced with available resources), and that the consumption vector x must lie in a closed convex set X in V+. (Convexity of X is forced by "rationality of consumer choices". Can you see why? Can you also see why it is natural to assume convexity and compactness of Y? closedness of X?) A price vector is a positive linear functional p on V, assigning a non-negative numerical value (in $$) to each combination of goods (with non-negative coefficients). Show that if X and Y intersect, there is a choice of x = y and a choice of a price vector p under which the economy is in equilibrium, in the sense that the constraints are satisfied and that the income p(y) of the producers cannot be further increased (with this choice of prices) without violating the constraints. (You can do this either with or without Hahn-Banach.) Next, do a slightly harder case. Instead of requiring x = y, require y - x = z in V+ (production exceeds consumption by a fixed non-negative amount, representing "savings"). Assuming that X + z intersects Y, show there is a choice of x, y, and p, for which the economy is in equilibrium.
9/6, 9/8.
(no class 9/4)
Ch. 5, Ch. 6 due 9/13: Lax, Ch. 5, Exercises 5 and 6 (p. 44) and Exercise 8 (p. 48); Ch. 6, Exercises 8 and 10 (pp. 61, 62)
9/11-9/15Ch. 6 (cont'd), Ch. 7 due 9/25: Exercise sheet on the Bergman kernel
9/18-9/22Ch. 8, Ch. 9 (selections)  
9/25-9/29Appendix A, the dual of C(X)  
10/2-10/6Ch. 10, Ch. 11 (selections) due 10/16: Exercise sheet on duality for Banach spaces, weak* convergence, and convergence of Fourier series
10/9-10/13Ch. 12, 13.1  
10/16-10/2013.2-13.4 due 10/25: Exercise sheet on Krein-Milman
10/23-10/27Ch. 14 (selections) and Ch. 15 due 11/15: Exercise sheet on distributions, the Fourier Transform, and linear operators
10/30-11/3Ch. 16 (selections) and Appendix B  
11/6-11/10Appendix B (cont'd) and Ch. 16 (cont'd)  
11/13-11/17Ch. 17, Ch. 18  
11/20-11/22
(no class 11/24)
Ch. 18 (cont'd) due 12/6: Exercise sheet on Banach algebras
11/27-12/1Ch. 19, Ch. 21, Ch. 22 (selections)  
12/4-12/8Ch. 28, Ch. 30 due 12/18: Final exercise sheet (cumulative)