Visualizing Data – Tracking Bubbles with Bubbles

Hal Varian has been spruiking the idea that the sexiest job in the world is that of the data scientist (in the New York Times). While we can only speculate as to whether Google’s Chief Economist is talking his own book, he makes a valid point – that new skills and techniques will be required to explore the overwhelming abundance of information in the ‘big data’ age.

Leaving aside the statistical task that the owners of big data have before them, the challenge is to define new ways of presenting information – both to express voluminous information in a coherent fashion and to explore the possibilities offered by our more interactive and dynamic technologies.

This is something that the financial media are slowly awakening to. Take, for example, The Economist’s house price indicator (click here) and Bloomberg’s overlaying of interactive graphics (for example, click here). Or simply read the advertisement for a graphic designer for Bloomberg’s multimedia group (here).

The great thing is that the democratizing power of the web means that we all can participate in the evolution – consider for example the following chart that I created with Google’s Motion Chart gadget:

I’d recommend taking some time to play with it – the original idea was to track bubbles with bubbles – try trailing individual countries by clicking on them or switching to a dynamic bar chart. The ability to easily remodel the variables being expressed in different dynamic formats hints at the possibilities.

Apart from enabling a glimpse at the functionality, the chart does suggest that market capitalisation relative to GDP has some value as an indicator. The logic goes that the types of companies listed on a country’s exchange are indicative of the composition of its economy. In aggregate, the earnings growth potential of these companies will limit the maximum sustainable market capitalization of the country.

The chart hints at this:

  • Try trailing only Japan and Italy – both countries with ageing populations, yet Japan with a higher technology component pushed through 100% market capitalisation to GDP in 1999/2000 and again 2006/2007.
  • An interesting addition to the analysis would be to add debt within an economy. Watching the inexorable climb of market cap to GDP for both the US and UK – incorporating debt might assist in understanding the relative accelerant provided by the credit bubble.
  • The resource-laden markets of Canada and Australia burned brighter than any other during the 2007 peak. They remain at historically expensive levels reflecting the emerging world’s demand for their dirt.
  • Isolating individual market sectors might also be an interesting addition to the analysis – highlighting when a specific sector may have gotten ahead of itself. For example, the financials in the US, UK, and well just about everywhere in 2007.

Conclusion

This type of analysis is likely to become more widely used and more accessible. That’s a good thing. Somewhat perversely though, such media probably requires more brainpower from its readers. The charts might make information more accessible, but they are still compressing more information – that is, after all, the point. The objective is to display the information in a way that enables our innate trend spotting skills to come to the fore. In this, we are still the undisputed kings (Take the Financial Turing Test here).

For further reading on big data and data visualization:

In-house Analyst
0

Comments

Leave a Reply

Your email address will not be published. Required fields are marked *