Friday 5 September 2014

I heard the news (analytics data) today

11:09 Posted by The Thalesians (@thalesians) No comments

It's perhaps a clichéd question to ask what is your favourite Beatles song? Ask most people and they'll reel off one of the well known tracks. Surely, "Yesterday" is one of these. Perhaps "Ticket To Ride" with that infectious guitar riff. Perhaps "Hey Jude", the piano chords waltzing in the background underneath Paul McCartney's vocals. However, one of my favourites is "In my life". It's a wonderfully simple track, only 2 minutes 27 seconds long, a track which most of you are unlikely to have heard.

I'm listening to it, whilst my fingers pulse the various keys on this keyboard, John Lennon's voice coming from the right of me, harmonies from Paul McCartney and George Harrison, punctuated by Ringo Starr drumming through my left speaker.Another Beatles track also seems to linger in my memory, namely "A day in the life", which has several contrasting sections. It begins with the line:

I read the news today, oh boy

It's perhaps a line which can resonate with everyone who follows markets. We hear news, we react. That's what we do in markets, we read the news and sometimes our reaction is simply "oh boy". The problem is that these days there is so much information from markets. Not only do we observe market data in its droves, but news data seems to multiply endlessly.

Before, maybe it was just newswires. Now the web is awash with blogs and other market moving commentary. Then we have Twitter too. It is impossible for a human to look at every bit of news and interpret it. The news gathering process becomes a matter of prioritizing what is "important" and what is "irrelevant" news. Sometimes however, news can slip beneath the your radar, only to suddenly tip the price action later. At other times, price action reacts almost immediately such as after payrolls data. The signal seems subsumed by masses of noise and can often be interpreted in so many different ways. Take for example the picture above, which I took a few months ago in a Budapest gallery. What I see in it, could be very different to what you observe. The same is true of markets. However, there is also a crucial difference in markets, there is only one "correct" interpretation, that of the price action!

So what can we do in this challenge to interpret news to extract that signal from in amongst so much noise? One solution is increasingly to look at news analytics. Data vendors essentially parse the news and then output summaries of news articles/blog posts in an easier to process format that can be read in a systematic fashion. One of the most well known is RavenPack, who have recently sponsored some of my Thalesians research, specifically to investigate how news analytics data can be used to trade markets. I found that news data could be used as a filter for FX carry (see results below - and ask me for a copy of the paper if you are interested, I am also offering a free subscription for shortened versions of Thalesians quant notes). Obviously, there are many other uses, but this is just a single example.


Compared with other data sources, news analytics is a relatively new area. Even amongst those who are aware of the idea, a lot of people can be dismissive. To some extent, this is good news if you want to trade off it! If a data source is relatively underutilised, strategies based on it, are unlikely to be as crowded. Furthermore, if it takes time to develop an algorithm to extract alpha from a data source, it also reduces the likelihood of crowding around a signal. Having spent a considerable amount of time analysing news analytics for trading purposes, it also seems apparent that there are numerous ways of trading on the data. Again a good point for reducing crowding around a specific strategy.

So next time you think you haven't got enough time to read the news, consider using news analytics. The last thing you want to be doing is reading the news, only to realise you've missed something and mutter "oh boy!".

To read more about ideas about using RavenPack news analytics data see my Thalesians quant paper listed below. My book Trading Thalesians also has some colour on reading through the noise when trading (mixed in with a bit of ancient history). Pre-order on Amazon UK here and Amazon US here.

Thalesians - Carry the news trade - Filtering FX carry using RavenPack news analytics - 2 Sep 2014 (Ask me for subscriber details! Shortened versions of my papers are available for free. Given this is a sponsored paper the full version is available free on request!)

0 comments:

Post a Comment