Matt, I think the individual pieces may be a great way to go. I was looking at SymPy's matrix code and was curious if that supports symbolic computation right now or is it only numeric? As for some of the individual pieces, I think the following are important concepts needed in financial statistics that I am unsure SymPy currently has functionality for: kernel density estimators for PDFs Empirical distribution functions Statistical moments Stochastic Processes Stochastic/Ito Calculus Time series Matrix decomposition (spectral, singular value)
Another option I haven't seen around is algebraic geometry. there's a lot of symbolic mathematics that can be done on curves/surfaces/higher dimensional objects such as the reduction of quadric surfaces, curvature, fundamental forms, etc. I am however most interested in the first set of topics, which I think are strong tools that are applicable to many fields. Best, Chase On Monday, March 3, 2014 6:18:05 PM UTC-8, Matthew wrote: > > Hi Chase, > > Thanks for your interest. Many of the topics you bring up are definitely > of interest to us. > > In my experience students who present and run with their own topics often > produce good results, so, if you have a particular interest in quantitative > finance then it's good to push on that. However, specific > application-focused projects tend to see a bit less use than general > mathematical infrastructure; they are only used by their domain rather than > re-used by lots of domains. I wonder, are there some interesting pieces of > mathematics on which finance depends that you could implement instead? In > other words is it possible to break down finance into various general > mathematical pieces, implement/improve those, and then finally cap the > project with a very thin finance layer? > > That being said, my understanding is that the Python/finance world is > pretty big and our current community doesn't currently make efforts to > support it in the same way that we do the sciences. Having someone around > who thought about how SymPy could benefit the financial world is probably > good for the project. > > -Matt > > > On Mon, Mar 3, 2014 at 4:43 PM, Chase Relock <[email protected]<javascript:> > > wrote: > >> Hi all, >> My name is Chase Relock >> I'm current a senior in mathematics at UC Berkeley and have experience >> programming python primarily. I've done two classes in pure group theory >> and have a lot of resources at my disposal on group theory that make it a >> viable choice for me. I've also done a high level ODE theory course which >> I've retained a lot of material from and could contribute. Ideally though I >> might be interested in implementing some statistical finance module. A >> built in black-scholes model and the calculation of implied volatility >> (Often something goal-seeked in excel) could be a place to start. I have >> also developed some interesting results from this question I initially >> asked on Quant StackExchange >> HERE<http://quant.stackexchange.com/questions/10407/algorithmical-repilcation-of-a-profit-and-loss-function-using-different-options> >> that >> allow for a very nice symbolic construction of a portfolio from a given >> payoff function that is more elegant than the solution proposed in the >> linked PDF. It would actually go hand-in-hand with a symbolic matrix >> library. This is a topic I will also most likely be writing a small paper >> about as I find that there is an interesting argument to be made about >> finding a minimum collection of options that generate a portfolio. >> Stochastic finance also very quickly leads to the usefulness of a >> stochastic process module. Please let me know if anything here would >> actually be of interest, as I'd be very excited to implement some of these >> ideas. >> >> -- >> You received this message because you are subscribed to the Google Groups >> "sympy" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to [email protected] <javascript:>. >> To post to this group, send email to [email protected] <javascript:> >> . >> Visit this group at http://groups.google.com/group/sympy. >> To view this discussion on the web visit >> https://groups.google.com/d/msgid/sympy/d492a9ed-c74a-4b66-8a8d-844241f4130b%40googlegroups.com >> . >> For more options, visit https://groups.google.com/groups/opt_out. >> > > -- You received this message because you are subscribed to the Google Groups "sympy" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/sympy. To view this discussion on the web visit https://groups.google.com/d/msgid/sympy/18faee70-57ea-4919-b2dd-a0093cc047f3%40googlegroups.com. For more options, visit https://groups.google.com/groups/opt_out.
