Dear All, Just wondering if some of the people who have either commercial experience in/or have studied quant finance can help me? I am currently working for a company that produces a commercial fluid dynamics engineering code and so have a strong background in numerical analysis, but it is more from a commercial point of view rather than fundamentals. Anyway, I got interested in quantitative finance a few years ago and so have been trying to get an entry-level position. For several years, I thought the main skills were the technology since I already have a strong grounding in numerical PDE solution, finite difference methods and some monte-carlo type techniques. However, recently I did a math test for a quant position and the maths blew me away !! So I have realised that I need to brush on my stochastic calculus as well as my tech skills. I have already got the Hull and Wilmott books and I intend to work through the questions in these. However, as part of the maths test I had to integrate the error function, differentiate integrals, etc and it has been a long time since I did this. This is also above the level that I think topics are covered in Hull,etc. I have already got an more advanced book on Stochastic calculus by M.Steele, and have ordered a few more such as Oksendal and will work through these, but all these books use various mathematical formulae that are covered in undergraduate/graduate textbooks. I wish to cover some of the basic required maths before getting into the nitty gritty of the Stochastic concepts and was wondering if anyone could give me some hints and tips of what areas I need to be looking at. This is what I have got so far...Calculus - Error function, Stirling's Formula, Gamma Function, Liebniz formula for differentiating integrals, Solution of differential equationsProbability and Statistics - Basic probability, finding expectation and variance functions for discrete and continuous distributions, binomial distribution, poisson and especially normal distributionSeries - Taylor series expansionsAnything I am missing that is especially important, or would people I recommend I cover these and then dive into the Stochastic calculus books ? Thanks in advance, Blade

HiI sounds like you picked up the practical stuff along the way and never learned it formally? most of the stuff you mention you would pick up on a first year math course, so I would check out some of the big universities webpages for math and pick some courses from there. you might even get some good notes. I would start of with analysis, linear algebra, differential equations (try for more theoretical stuff) then dive into some of your stochastic books (etheridge is a good one), wilmott, hull .. also try 'the mathematics of financial derivatives' by wilmott/howison/dewynne because you might be able to pick that up straight away. Also a good book on basic probability won't hurti wouldn't worry too much about that test though... there could be a job out there for you, and not everyone will give you questions like that!

Thanks MonkeyA, I went to Imperial college and did an MEng in Engineering so did alot on differential equations, taylor series expansions etc. but never much on probability and statistics. I have really had to start from undergrad level on this subject and am learning slowly. However, it has been three years since I graduated and there is alot you forget. So for the calculus side there will be some revision but I really will be learning from scratch on the probability and statistics side. What I wanted to know is whether I needed to really look at 1) Permutations and combinations2) Regression and correlations3) Integration of distributions other than the normal distribution, and if so what are these distributions ? I already have the 'The mathematics of financial derivatives' so I think once I have covered the the basics of probabilty and statistics I should get cracking on that. Also, thanks for the recommendation of Etheridge, I haven't heard this mentioned before and will look into it. I guess you are right in the fact that I shouldn't let this one test throw me though. I already work in a very mathematical area and I look up most of the stuff I need instead of remembering it so I should probably focus on learning the basic methodologies rather than specific mathematical relationships. Thanks, Blade

One place to look for courses is MIT's Open CourseWare project.. They have very good lecture notes, syllabi, exams, etc. for a large number of investment/business and math topics.Daniel

Yes the OCW site of MIT is really good for that sort of thing.blade... I don't think you need to know any of the three topics you mentioned right now!Your knowledge of (i'm guessing...) numerical techniques for solving ODEs, PDEs etc. (in C++?) will mean you are good for a job involving pricing of options, developing and testing models etc.... the things you are looking at would be more for people doing stat arb, volatility modeling etc. I think you need to read 'the mathematics of financial derivatives' right now for some motivation! Take a look at: G. R. Grimmett and D. J. A. Welsh, Probability: An Introduction Its an easy to work through book, and get you the ideas on probability you are after... p.s. I know a few people who graduated imperial three years ago ... maths and biotech people mainly.

Hi MonkeyA, Yes, I have good experience of numerical solution of ODE's and more specifically PDE's but I am not a hardcore developer. Unfortunately we work in fortran, if I had done C++ then I would have probably walked into a techie quant role by now. I have learnt C++ and wrote a code to solve the Black-Scholes PDE ( from Wilmott) using this language but the main problem I get is that people need commercial C++. I have already been offered development roles within my company but I am not keen to work in fortran. Apparently we are making new codes that use C++ so maybe I will get the chance. In terms of the maths, I thought that this was the sort of knowledge that people would require Structurer - formulating models for new products, volatility modellling, stat arb and they would need fundamental mathsDeveloper - since Black-Scholes is already hedged - more knowledge of the numerics of ODE's, PDE's, how equations behave when you discretise them, and optimisation of code - This is the area I have most experience in. Trader - Knowledge of how the code behaves, the greeks, experience in applying in different techniques to different markets. I am currently working on a part of our code in which we model Brownian motion ( we call it turbulent dispersion) and I am having to look at alot of these statistical techniques and various transition density functions ( Fokker-Planck equation ) since we need to improve the model. So I am getting some practical knowledge of these stochastic concepts as well. Btw. I graduated in 2000 - I knew quite a few people from maths and a few from Biochem, I did aeronautical engineering myself .... What's your background ?