M3/4/5F22 Mathematical Finance N. H. Bingham Imperial College London, 6 October - 15 December 2017
Uploaded by
Sylvia CheungM3/4/5F22 Mathematical Finance N. H. Bingham Imperial College London, 6 October - 15 December 2017
Uploaded by
Sylvia Cheungm3f22ch0.tex Chapter 0 6.10.
2017
M3/4/5F22 MATHEMATICAL FINANCE
N. H. BINGHAM
Imperial College London, 6 October – 15 December 2017
6M47; 020-7594 2085; n.bingham@ic.ac.uk; Office hour Mon 3-4
Course website: My homepage.
Room 340; Mon 4-5, Tue 9-10, Fri 9-10.
Books.
Course text: Ch. 1-6 of
[BK] N. H. BINGHAM and Rüdiger KIESEL: Risk-neutral valuation: Pric-
ing and hedging of financial derivatives, 2nd ed., CUP, 2004 (1st ed. 1998).
Alternatives:
S. E. SHREVE: Stochastic calculus for finance. Vol. I: The binomial asset
pricing model; Vol. II: Continuous-time models, Springer, 2004.
T. MIKOSCH: Elementary stochastic calculus, with finance in view, World
Scientific, 1998.
Course Website: M3F22 link on my home-page (Imperial College > Mathe-
matics Department > Staff > Staff List > Bingham > Homepage – or Google
Nick (or Nicholas, or N H) Bingham: favouritize this to get it in one click).
Other relevant links (on my home-page):
[SP] Stochastic Processes [30 hours, MSc, Mathematial Finance];
[SA] Stochastic Analysis [20 hours, MSc].
[LTCC] Measure-theoretic probability theory [10 hours; MSc].
For background:
[PfS] Probability for Statistics;
[SMF] Statistical Methods for Finance;
[Math482] – a course along these lines I gave at Liverpool.
Mathematical finance, for reference
[CR] John C. COX and Mark RUBINSTEIN: Options markets. Prentice
1
Hall, 1985.
[H1] HULL, J. (1995): Introduction to futures and options markets (2nd ed),
Prentice-Hall, (‘baby Hull’), or
[H2] HULL, J. (1993): Options, futures and other derivative securities (2nd
ed.), Prentice-Hall (‘Hull’).
Background and general interest
[K1] John KAY, The Truth about Markets: Their Genius, their Limits, their
Follies. Penguin/Allen Lane, 2003.
[K2] John KAY, Other People’s Money: Finance: Masters of the Universe or
Servants of the People? Profile Books, 2015, £16.99).
[AH] Anat R. ADMATI & Martin HELLWIG, The Bankers’ New Clothes:
What’s Wrong with Banking and What to Do about it, Princeton UP, 2013.
[G] Alan GREENSPAN, The age of turbulence. Penguin, 2007.
I think the Kay books are essential reading for anyone thinking of working
in the financial services industry – maybe [AH] too. I thoroughly recommend
[G] – but get the latest edition of it that you can. The author was Chair-
man of the US Federal Reserve (Fed) 1987-2006. His views up to 2007 were
largely Panglossian optimism (markets know best, and are self-correcting,
etc.). The ongoing problems since have forced a re-think; see the epilogues
to later editions, his evidence to the House Committee, etc.
Mathematics, for reference
[BF] N. H. BINGHAM and J. M. FRY: Regression: Linear models in statis-
tics. Springer Undergraduate Mathematics Series (SUMS), Springer, 2010.
[D] J. L. DOOB: Stochastic processes, Wiley, 1953.
[N] J. NEVEU: Discrete-parameter martingales, North-Holland, 1975.
[KS] KARATZAS, I. & SHREVE, S. (1988): Brownian motion and stochas-
tic calculus. Graduate Texts in Math. 113, Springer.
[RY] REVUZ, D. & YOR, M. (1999): Continuous martingales and Brownian
motion. Springer, 3rd ed. (1st ed. 1991, 2nd ed. 1994,).
[RW1] ROGERS, L. C. G. & WILLIAMS, D. (1994): Diffusions, Markov
processes and martingales, Volume 1: Foundation, 2nd ed. Wiley.
[RW2] ROGERS, L. C. G. & WILLIAMS, D. (1987): Diffusions, Markov
processes and martingales, Volume 2: Itô calculus. Wiley.
Mathematics for reference: Insurance Mathematics
[A] S. ASMUSSEN, Applied probability and queues, 2nd ed., Springer, 2003
2
[1st ed. Wiley 1987].
[AA] S. ASMUSSEN and H. ALBRECHER, Ruin probabilities, 2nd ed.,
World Scientific, 2010 [1st ed., S. Asmussen, 2000].
[Kyp] Andreas E. Kyprianou, Fluctuations of Lévy processes with applica-
tions: Introductory lectures, 2nd ed., Springer, 2014 [1st ed. 2006].
[RSST] T. ROLSKI, H. SCHMIDLI, V. SCHMIDT and J. L. TEUGELS,
Stochastic processes for insurance and finance, Wiley, 1999.
Assessed Coursework: One assignment, 10% credit, Week 6 (due Week 7).
CONTENTS
I. PROBABILITY BACKGROUND [c. 2.5h]
§1. Area: prelude to measure
§2. Basics: recapitulation from Years 1 and 2
§3. The Poisson process; compound Poisson processes
§4. Transforms.
II. ECONOMIC AND FINANCIAL BACKGROUND [c. 4.5h].
§1. Time value of money; discounting
§2. Economics and finance; utility
§3. Brief history of mathematical finance
§4. Markets and options
§5. Portfolios and hedging
§6. Arbitrage
§7. Put-call parity
§8. An example
§9. Complements
§10. Postscript to Ch. II. Systemic aspects – ”Big-picture stuff”
III. MEASURE-THEORETIC PROBABILITY [c. 3h].
§1. Measure
§2. Integral
§3. Probability
§4. Equivalent measures and Radon-Nikodym derivatives
§5. Conditional expectations
§6. Properties of conditional expectations
3
IV. STOCHASTIC PROCESSES IN DISCRETE TIME [c. 2.5h].
§1. Filtrations and information flow
§2. Discrete-parameter stochastic processes
§3. Discrete-parameter martingales
§4. Martingale convergence
§5. Martingale transforms
§6. Stopping times and optional stopping
§7. The Snell envelope and optimal stopping
§8. Doob decomposition
§9. Examples
V. MATHEMATICAL FINANCE IN DISCRETE TIME [c. 5h].
§1. The model
§2. Viability: existence of equivalent martingale measures (EMMs)
§3. Complete markets: uniqueness of equivalent mg measures
§4. The Fundamental Th. of Asset Pricing: Risk-Neutral Valuation
§5. European options. The discrete Black-Scholes formula
§6. Continuous-time limit of the binomial model
§7. More on European options
§8. American options
§9. American options: Infinite time-horizon.
VI. STOCHASTIC PROCESSES IN CONTINUOUS TIME [c. 4h].
§1. Brownian motion
§2. Filtrations; finite-dimensional distributions
§3. Classes of processes
§4. Quadratic variation (QV) of Brownian motion; Itô’s Lemma
§5. Stochastic integrals; Itô calculus
§6. Stochastic differential equations (SDEs); Itô’s Lemma
VII. MATH. FINANCE IN CONTINUOUS TIME [c. 4h].
§1. Geometric Brownian motion (GBM)
§2. The Black-Scholes model
§3. The (cts) Black-Scholes formula: derivation via Girsanov’s theorem
§4. Related results.
§5. Infinite time-horizon; American puts
§6. Real options (Investment options)
§7. Stochastic volatility (SV); Postscript to Math. Finance
4
VIII. INSURANCE MATHEMATICS [c. 4.5h]
§1. Insurance background.
§2. The Poisson process & compound Poisson processes (ctd).
§3. Renewal theory.
§4. The ruin problem.
§5. Lundberg’s inequality.
§6. The ruin problem and the renewal equation.
§7. Cramér’s estimate of ruin.
§8. More on insurance; Postscript to Insurance.
Exam
Exam in the summer. Standard format: M3F22, 4 questions; M4F22/M5F22,
five questions; Q5 Mastery Question.
About the course.
The course is an introduction to mathematical finance and option pric-
ing – Black-Scholes theory via martingales. The Black-Scholes formula (of
1973, which enabled the pricing of options on an industrial scale) changed
the world, and is certainly the most important new formula of the last part
of the last century. It is basic to the financial services industry. So: do the
course either for academic interest, or if you have any career interest in the
financial services industry – or both.
This is not a get-rich-quick course (if you wanted to get rich quick, why
did you choose mathematics?).
On the ‘finance’ side, you will learn something about Economics (below);
this is very important, and will be new to many of you. On the ‘mathemat-
ical’ side, you will learn stochastic (or Itô, or martingale) calculus. Calculus
is the most powerful weapon we have – in mathematics, or in science gen-
erally. You learn it at school, where it is a revelation; you re-learn it doing
Mathematics at university (less fun at first, for most people). The first obvi-
ously new and obviously powerful kind of calculus you learn here is Complex
Analysis in Semester 4. Think of stochastic calculus as a sequel to that.
About the course: Content.
There are two things here, one pure mathematical, one non-mathematical.
(a) Measure Theory.
‘Grown-up probability’, needed here (Itô calculus, etc.) is measure-theoretic.
5
Ideal preparation would be a full course in Measure Theory (such as M3P19
Measure and Integral, Autumn Term), and then Probability (M3P6, Spring
Term). But only a minority of students attending this course will have had
these. So, we deal with necessary measure-theoretic preliminaries in Chapter
III. In the time available, one cannot prove the guts of technical measure
theory – the key approximation arguments. So we quote these, confining
proofs to what goes before and what comes afterwards (both much easier).
(b) Economics.
Finance is a small and specialised part of Economics. Ideal preparation
would also include a good grounding in Economics. There would not be room
for this in the Maths curriculum here, and most of you will not have an Eco-
nomics qualification from school. So again, we have to take a lot for granted;
we cover the necessary economic and financial background, in Chapter I.
In this regard, please bear three things in mind:
1. Anything important enough becomes political (M. Maurice Couve de
Murville). This stuff is certainly important.
2. Politics in not an exact science (Bismarck). But,
3. Mathematics is an exact science.
We will be doing lots of mathematics – in particular, we derive the Black-
Scholes formula. We will extend calculus, the most powerful single weapon
we have, to become probabilistic (Itô calculus) and apply it to these prob-
lems. But, there are limits to which finance, economics, or anything involving
human psychology, is mathematicisable. As always in Applied Mathematics,
we have to be on guard: if we don’t simplify enough, we can’t do anything;
if we over-simplify, we can do things, but can’t trust our conclusions.
Just as important as the technical mathematics, you need to think about
the systemic faults at the geofinancial/economic/political level thrown up by
the crisis of 2007-08 on (Credit Crunch, etc.). Any prospective employer in
the financial services industry should ask you questions about this, and your
views on it, in interview. We spend half of one lecture on such things (II.10,
– not examinable). The rest is down to you.
For a range of views here, see e.g.
Quantitative Finance 15 no. 4 (2015), Special Issue on Interlinkages & Sys-
temic Risk, esp. Dempster’s review of the Admati-Hellwig book, 579-582;
N. H. BINGHAM: The Crash of 2008: A mathematician’s view. Significance
5 no. 4 (2008), 173-175 [on my home-page, under Papers]. NHB