Archive for January, 2010


Media Still Blaming the Quants

WSJ journalist Scott Patterson has jumped on the bandwagon suggesting that quants where a major cause of the GFC...

Conventional wisdom says that quants formulate models in conjunction with traders. Once developed they tell the traders how the model internals works, pointing out assumptions and limitations so that an informed opinion can be formed about when the model should be applied and when it's not going to work. Quants should be able to quantify the risk and expected loss under certain "bad" scenarios.

When the trading model is in operation, further up the chain back-office track exposures and risk, and at the very top bank executives have overall control of the allocation of capital to trading desks. Thus there are many levels of accountability: the quant, the traders, back-office, and executives. If people other than the quants don't understand the models and their application then they ought to be very caution about applying then.

I don't see that as a failure of quants. Sure, not all models are well thought out but in my mind it's a failure of risk management, or a strong preference by bank executives for profit over prudent risk management that is more to blame than the quants.


Negative Numbers using Two’s Complement

A colleague asked me about question (15) in my Uber Geek Challenge post so I thought I'd provide a simplistic explanation as I don't think Wikipedia gives such.

Basically, there are two popular ways of storing negative numbers in a binary number system: one's complement and two's complement. The easiest to understand is, arguably, one's complement. To get the negative of a given binary value when using one's complement you simply flip all the bits. However with zero represented by all zeros in binary, flipping all bits results in all ones and this is interpreted as "-0". Having two values for zero (one positive and one negative) isn't desirable so the alternate approach, two's complement, flips all the bits then adds 1. That ensures you don't get a positive and negative zero, and the only price you pay is a slight asymmetry in the numeric range - one less positive number if you don't count zero as positive.

The image on the right shows the binary representation of 8-bit negative numbers and their positive counterparts.


Innovation Takahashi Style

Recently I was kicking around some ideas with some work colleagues on the topic of software development and architecture (my slide deck is included below). On this topic I'm a firm believer in taking a holistic view on software development. That is, you need to focus not only on the details of the technical elements but also on both the short- and long-term business objectives of the process. This applies to the technical team as well as the BAs/managers driving the project.

More often than not there are both functional goals and business goals attached to the development exercises. The functional spec, if you have one, will elaborate on the functional goals/reqs that perhaps you have mutually agreed with your client/customers, or you may have crafted them without input from your current customers. This identifies what you are going to build but often doesn't indicate why you are doing it. Sure, "this feature solves that problem for the user" might sound like the obvious answer, but resource constraints force teams/companies to only do a limited amount of things that they could do. So then the question becomes what to prioritize, given your business' strategic objectives. These strategic goals are especially significant if the revenue of the firm is dependent on the quality and market penetration of the software you are building. After all, you need to ship before you can sell but once it hits the street [after countless iterations no doubt] your product needs to stand tall next to competitors' products. That brings me onto the topic of innovation...

Why is innovation important? Well, if you just imitate others you'd better out-execute them, or under-price them because that's all you'll have on them. This works for some companies but not for long and the margins are never great. The harder but significantly more profitable route is to out-innovate your competitors by building features that add a lot of value to the user and ARE HARD TO DO. Things that are easy and useful are going to be done by everyone - but the things that are very hard to do present a barrier to innovation that many organisations can't address even if they recognize the issue. So go looking through your Too-Hard basket and see if there are seemingly-crazy, but potentially killer ideas in that lot. You might be surprised!