[Quantlib-dev] Betonmarkets.com / Quantlib

[Jean-Yves S. <jy...@re...>]

Dear Quantlib developers,

I would like to introduce our company Betonmarkets.com.  Our website
specialises in fixed-odds financial bets (aka digital options).  We
have offices in the Isle of Man, Malta, and Malaysia, and were founded
in 1999 as an affiliate of Regent Pacific Group, a Hong Kong-listed
company.  Our website is currently the largest fixed-odds financial
betting website, transacting in over 20,000 bets per day.

Currently our bet pricing is performed in-house using our own
algorithms.  We are considering the feasibility of migrating to
Quantlib, and the purpose of this email is to request feedback as
regards the feasibility of such a move.

Due to the nature of our website - i.e. the ability to price digital
options and get an instant response - our bet pricing model needs to be
fast (response time under 1 second), which rules out computationally
intensive numerical methods, unless they could be accelerated by using
approximations or assumptions such as a simplified model for the vol
surface.

Our website offers a wide range of digital options, including digital
call, put, one-touch, no-touch, range (all of these in both European and
American flavours), as well as more exotic digitals such as 'up or
down', 'expiry miss' and others.

We'd be most grateful for your thoughts on the feasibility of adopting
Quantlib, which pricing models might best suit our purposes, and what
optimisations/approximations we may think about implementing in order
to decrease computational intensity.

Kind regards
Jean-Yves Sireau

-- 
Jean-Yves Sireau
CEO, Regent Markets Group Ltd.
Genseq Ltd.