Words & numbers in trading

There are words and numbers. It seems drummed into us from an early age that we're either a "words" or a "numbers" person. Perhaps, it is an over generalisation, to say that there's a certain pride at not being good at maths if you're "words" person, or having poor spelling if you're a "numbers" person. Of course, we could complement our list with "art" and many other skills. Whilst, I enjoy writing a lot, I have to confess I'm probably more of a "numbers" person than a "words" person if I was forced to make a clear decision.

However, it wasn't always like this. Going back in history to the ancient world, there were many famous polymaths, such as Thales. His expertise spread across many disciplines, philosophy, mathematics and astronomy, naming just a few. Indeed, one of the chapters of my book Trading Thalesians discusses the idea that having a wide array of interests like Thales can be very positive for modern day trading. As for "art" and "numbers", figures such as Da Vinci, remind us that these subjects need not live in totally different spheres.

Markets are becoming ever more complicated. Information is flowing from many different sources. Simply relying on one sort of data or approach to understand markets can end up missing out a big part of what's going on. Numbers are of course the crux of any sort of algorithmic trading strategy, we cannot get away from that. Even if we are a discretionary trader, who uses human judgement to make trading decisions, a modicum of quantitative analysis can help us. 

One of the biggest events of the coming weeks is the UK General Election. Much has been written about who could win. Strategists have outlined the potential outcomes. Of course, a lot of the commentary could be biased! As with art, there are many ways to see the same situations, zoom in and see something, zoom out and get a totally different interpretation!

What about trying a quantitative approach for analysing the UK elections, using numbers to compliment a more fundamental "words" take? I recently wrote a paper for the Thalesians (Thalesians: My kingdom for a vote), which attempted to do this. The idea was simply to understand how markets had behaved in historical elections, depending on the winner. I also analysed how well polls predicted the results. Interestingly, since 1974, on most occasions, the largest party was in line with polls prediction. The election of 1992 was a major exception. It surprised pollsters, although in that instance, the polls were very close beforehand. However, even more surprising is the way GBP/USD has behaved historically following UK general elections. In most cases, whether there was a Conservative or Labour victory historically, GBP/USD depreciated. We can see this in the chart below (the "art" bit!), which I have taken from my paper.

(If you like coding up Python, I've written a blog post at Plot.ly, which contains the Python code for the analysis I did for the chart below, as well as how to make it interactive using Plot.ly! You can view the below plot interactively here too.)

Of course, I am not advocating that we should always forget about words when analysis markets... indeed, we can convert words into numbers too, so "numbers" people can understand "words"! Over the past year, I have done a significant amount of work looking at news data, which is converted into a machine readable format together with a sentiment score and other metrics that attempt to quantify news. I have found news data to be very useful for a systematic approach to trading, doing projects for RavenPack, a news data vendor, showing how it can be used to reduce drawdowns of FX carry and improve upon long only bond futures (those papers are available on request). The key of course though is to have a well thought out framework for how you wish to use news data in your trading, before undertaking the analysis.

If what I've talked about sounds interesting and you want to learn more about how to be a "numbers" and a "words" person when it comes to systematic trading, come to one of my talks. I'll be speaking about both news based trading and also volatility markets during scheduled events at Global Derivatives in May in Amsterdam. I hope to see you there!

Why be either a "numbers" or a "words" person when you can be both!

Like my writing? Have a look at my book Trading Thalesians - What the ancient world can teach us about trading today is on Palgrave Macmillan. You can order the book on Amazon. Drop me a message if you're interesting in me writing something for you or creating a systematic trading strategy for you! Please also come to our regular finance talks in London, New York and Budapest - join our Meetup.com group for more details here (Thalesians calendar below)

17 Apr - Budapest - Impact of bitcoin - Tamas Blummer & Panel featuring Izabella Kaminska / FT)
22 Apr - New York - How Smart Money Invests and Market Prices Are Determined - Lasse Pedersen
29 Apr - London  - Global macro & UK election panel - Eric Burroughs / Reuters, Mark Cudmore / Bloomberg, Jordan Rochester / Nomura, Jeremy Wilkinson-Smith / Independent & Saeed Amen / Thalesians