I was in Boston last week attending the ASME International Mechanical Engineering conference demonstrating MapleSim, our new tool for physical modeling.  I had the opportunity to speak to a large number of delegates, but I remember one conversation in particular; a professor who taught freshman students was bemoaning the fact that he found it harder and harder to impress students with his relatively simple animations of physics phenomena.  A simple animated pendulum no longer captivated students who were already immersed in the interactive physics-enabled environments of video games.  He had to escalate the intricacy of his demonstrations, but generating them was starting to consume too much of his time.

Twenty years after I first plotted the Mandelbrot set on a ZX Spectrum with 48K of RAM and a 3.5MHz processor, I’m still amazed by the sheer complexity and beauty contained therein.  I now have access to far more computing horsepower and can create ever more vivid visualizations.  It’s surprising what you can do with some creativity and a modicum of patience.

Like most students studying engineering in the 90s, spreadsheets were the de facto calculation tool.  I used them for everything from food budgeting to pump and piping sizing calculations. 

Computing power has since exploded, and engineers have far better choices.  But engineers still continue using spreadsheets. 

Why?

There’s really only one reason – ubiquity and familiarity.  A spreadsheet is installed on nearly all desktop computers, but even though most engineers are aware of at least some of their design deficiencies they keep on using them.

That’s a mantra I need to have drummed into me, and perhaps tattooed on the inside of my car so I’m reminded every morning.  But I keep on making the same mistakes. 

 I seem to think that if I’ve “optimized” my portfolio with a few flashy calculations that I’ve done my due diligence, and the next stop is financial independence.  It’s the black box syndrome – trusting the output of a computer program without truly understanding the real issues.  Most portfolio analyses, for example, hinge on historical data, which of course doesn’t predict the sub-prime blow-up in the US or whether Brazilian coffee growers are on strike.  They’re all backward looking.

 However, in the absence of a neighbourhood scryer, historical analyses are a good indication of how to position a portfolio for the long term.

 Being a geek (however much I strenuously deny it), I tend leverage my tech skills wherever I can.  I wrote the attached worksheet to import stock quotes, including historical data, from Yahoo using the Sockets package.  Simply type in the appropriate NYSE stock tickers into the appropriate text boxes, check the quantities you want to download, and click the big gray button.

 All the stock quotes and historical data can be manipulated on the command-line and can be accessed via command-completion. 

 It then finds the best distribution of stocks in a portfolio by maximizing its Sharpe Ratio (through the Optimization package). 

The Sharpe ratio quantifies how effectively a portfolio of risky assets utilises risk to maximise return.  It’s defined as follows.

It essentially measures how effectively a portfolio uses risk to maximize return – the higher the ratio the better.  The expected portfolio return is predicted from historic data, the portfolio standard deviation is traditionally used as a proxy for risk, and the risk free return is the return that can be expected from a zero-risk investment (i.e. the interest on US Treasury Bills or the redemption yield on UK gilts).

What I find particularly fascinating is how Maple is now the centre of my technical desktop.  Through the combination of the interface and its math tools, I now use it for everything from the simplest calculations through to making wild guesses about my financial future.  If any of the developers are reading this, I want you to know there’s a lot riding on you...