No honour amongst code thieves August 19, 2009
Posted by James Webster in : finance, development , add a commentIts interesting that there have been a few stories about theft of code relating to trading systems lately…
The first to break in July was the story of ex-Goldman Sachs developer Sergei Aleynikov taking algorithmic trading code from his former employer. This event also seemed to be the impetus for greater scrutiny of the high-frequency trading that Goldman Sachs and many other Wall St players use to get their edge and increase alpha, albeit at the expense of retail traders. This prompted some exchanges/ECNs to reconsider their policy surrounding flash orders.
The second story broke in August; an Australian derivatives trading firm Optiver is suing a newly established rival, Tibra, over allegations that the trader who left the former to form the latter took core trading software at the heart of Optiver’s competitive advantage.
The significance of these stories is that they highlight the ‘technological arms race’ that pervades the professional end of certain markets (equities in particular). Whether these alleged thefts actually could benefit their recipients is arguable; however the increased information security regulations that will be imposed on their former colleagues are likely to have a negative impact on productivity and collaboration.
Wolfram Alpha and mathematical finance May 20, 2009
Posted by James Webster in : web, finance , add a commentWolfram Alpha has been released to a mixed response; some people have been overly critical (plenty of discussion indexed by Techmeme), but us geeks undeniably love the idea even if the implementation is a little flakey; I’ve seen plenty of ’server too busy’ type messages, it renders horribly in IE6 (don’t ask…) and I’m not so keen on the use of images to render the text. Clearly the latter is to defeat screen scraping (some people have said they have seen the text in the ALT tags, but I’ve not encountered that myself) but maybe a technique like sIFR would be a better approach (without using the replacement text side of sIFR). I’m not sure how it will be monetised, or even if Wolfram plans to monetise it although it can’t be cheap to run. Possibly a paid API is waiting in the wings?
I’m impressed with the results of ticker queries, especially when combining multiple tickers; e.g. GOOG, AAPL, MSFT. I am even more impressed with the results for the following queries: ‘black scholes’ brings up a basic option pricer with all the parameters you might expect (’option’ and ‘option valuation’ also return an identical result) plus a standard payoff diagram, first-order greeks, a ladder for different strikes, all sorts of neat graphs showing the relationship between the greeks and other input parameters PLUS a little bit on how to statically hedge the option. Other searches such as ‘butterfly option’ or ’straddle option’ show the payoffs from these option strategies. I thought that ‘AAPL option’ might show the option series available for AAPL (or any other stock for that matter) but it looks like there is still work to be done there. Trying queries for more exotic types of options (e.g. asian, barrier, basket) was not successful. Still, it is definitely possible to see how this might evolve to compete in some small way with Bloomberg’s analytic tools as I commented upon last month (in that post I also mentioned Stocktwits; they have recently raised capital in a VC 1st round and are planning a premium version).
Continuing my EC2 theme from yesterday, there was talk from Wolfram at the end of last year that Mathematica would be available in some way on EC2, probably as a pay-per-use AMI. Although there was plenty of press coverage at the time (indexed by Wolfram here, search page for Cloud Computing) it still hasn’t been released as far as I can see… I wonder how Alpha will overlap with any Mathematica-on-EC2 offering.
On the topic of Mathematica, Patrick Collison has an interesting blog post about combining Mathematica with the Google Analytics API.
The Long Decline of Bloomberg? April 29, 2009
Posted by James Webster in : finance , add a commentThere seems to be a theme running around the financial/technology blogosphere at the moment, namely that there are a number of trends and startups potentially threatening the hegemony of Bloomberg and its eponymous terminal.
Etrading asks: Wolfram Alpha: another nail in Bloomberg’s coffin? and speculates that with a few key features (most notably availability of financial data) the hotly anticipated ‘analytical search engine’ could be a serious challenger to the charting/analytics tool sets in Bloomberg.
Zerobeta states: Bloomberg Needs to be Less Swiss Army Knife and More Leatherman and makes an argument that Bloomberg should focus on its core competency, financial journalism, embrace an open access approach to its data and allow an ecosystem to evolve around its platform.
Techcrunch reported on the launch of SkyGrid a financial news aggregator. Even their design (white text on black) looks suspiciously like Bloomberg.
Clearly the scope, depth and market penetration of Bloomberg far outreaches any of these potential competitors (plus others such as StockTwits and Chart.ly). Still with Bloomberg charging fees as high as it does (described as absurd by Zerobeta) and the competitive landscape continuously changing underneath it they will have to keep innovating (and to be fair, their iPhone app is great) to keep their dominant position.