# Where to begin in approaching Stochastic Calculus?

I have experience in Abstract algebra (up to Galois theory), Real Analysis(baby Rudin except for the measure integral) and probability theory up to Brownian motion(non-rigorous treatment). Is there a suggested direction I can take in order to begin studying stochastic calculus and stochastic differential equations?

-
When you say probability, is it the measure theory approach or the classical one? – Jean-Sébastien Nov 6 '12 at 22:20
most likely measure theory approach, considering this is stochastic calculus I imagine it would have to use measure theory. – Lee Jacobs Nov 6 '12 at 22:23
What I meant was, what kind of probability have you done – Jean-Sébastien Nov 6 '12 at 22:23
classical, It was non-rigorous. I have some idea of the measure theory approach, having worked with (again non-rigorous) martingales in a math finance course. You have a probability space and essentially probability is just a function that maps an event (or set of events) to a real number. – Lee Jacobs Nov 6 '12 at 22:28

I Suggest

For Measure Theory

1. Real Analysis -Royden
2. Measure Theory- Halmos.

For probabiltiy theory, Brownian motion and stochastic Calculus

1. "Probability with Martingales" by David Williams.
2. "An Introduction to Probability Theory and Its Applications 1-2" Wlliam Feller.
3. "Diffusions, Markov Processes and Martingales:1-2" by Chris Rogers and David Williams.
4. "Introduction to Stochastic Integration" by K. L. Chung, R.J. Williams
5. "Stochastic Differential Equations: An Introduction with Applications" by Bernt Øksendal.

You may also need to learn some Complex Analysis. Although Complex analysis is not essential to learn probability theory and stochastic processes. However, contour integration and Fourier transforms are indispensable tools and it is also one of the most beautiful and useful areas of mathematics.

-

As previous answers have indicated, a solid measure-theoretic approach to probability is essential. A book I would strongly recommend for the measure-theoretic approach to probability is:

R. M. Dudley - Real Analysis and Probability

I found it very clear, well-organised and invaluable when I was working on such things.

-

I would suggest to first take a course and/or get a book on probability theory, in view of measure theory. It helps to have studied measure theory first but is not necessary.

Some textbook ideas, but in no way an exhaustive list:

• Durret: Probability and examples. I've followed a class using this book, you can find an earlier edition online I think
• Billingsley: Probability and measure

I have not worked with the last 3 but only consulted for references. Perhaps our more experienced user could have a better say in the matter.

Once you have done that, you can take a class on stochastic calculus in general. That should explore the construction of Brownian motion, the Ito integral, some Stochastic Differential equations and a continuation of martingales that you will have started in course 1. Some books are

• Shreve, and also Steele have books with some financial emphasis
• Karatzas' Brownian Motion and Stochastic Calculus has been around a while but might be harsh for a first class

You can then take more advanced class on specific topic such as Stochastic Differentiual equations. One book that comes to mind is Oksendal's

-

In my experience it is best to get to your goal as quickly as possible to maintain momentum. You could read Royden and then Billingsley and finally start on stochastic calculus. But personally, I would probably run out of steam before long.

Jeff Rosenthal's book A First Look at Rigorous Probability.

It is 200 pages long. It is very clearly explained (baby Rudin is all you need). It develops all the measure theory you need in a probability context. It has a lot of easy exercises that build confidence that you understand basic concepts. Half of these have solutions!

The last chapter whets your appetite for stochastic calculus and he gives suggested reading.

-
Rosenthal's book actually reads very well. I like it. Thanks for the suggestion. (amazon preview) – Lee Jacobs Nov 7 '12 at 17:48