Expert discussion on the use of Big Data: “We want to keep the information edge on unstructured data”
In conversation with Markus Ebner, Head of Multi-Asset at Quoniam Asset Management.
Dr Ebner, it is often said that data is the new oil or the new gold. Would you go one step further and say that big data is even more valuable?
Yes, I would share this assessment. The amount of data has become much larger in the past few years, and at the same time the availability has improved greatly. And a lot can be done with this data, just think of targeted web advertising. But also in asset management, data sources and content are now available that were not accessible a few years ago at all.
How does this change your daily work?
As a quantitative manager, we used to take sources such as Datastream or Bloomberg to generate forecasts for individual stocks or markets. What is new now, is structured access to so-called unstructured data, such as news articles. Here, more and more data providers are doing the preliminary work, i.e. they filter the news according to certain terms. We then benefit from this pre-selection.
“Unstructured data are usually much more up-to-date than conventional research data, so we can capture market sentiment almost immediately.”
Dr Markus Ebner
Head of Multi-Asset
What are the advantages of unstructured data for asset management?
The data is usually much more up-to-date than conventional research data, so we can capture market sentiment almost immediately. After all, the goal is to get information that is not yet priced in. We have found that forecasts of certain markets, such as Spain, improve when we include sentiment factors based on unstructured data. This gives us a better understanding of the current economic situation and its future development. And this forms the basis for medium-term investment decisions.
But looking at the sheer amounts of raw data. How do you evaluate them and, above all, how do you assess what is suitable as an investment signal and what is not?
We currently work with two data providers, one is RavenPack, the other is GDELT, both of which monitor media worldwide from different perspectives and sort articles according to topics, countries, company names and the like. Depending on the asset class, we have aggregated different topics into which we classify the data. From this we build time series and can see how sentiment is developing per country and asset class. And from this we then derive trading signals.
What is innovative about Quoniam’s approach? What do you do different compared to other managers?
Our strength at Quoniam is identifying relevant topics and then deriving trading signals from them. We have built up a wealth of experience over eight years, which distinguishes us from other providers. Apart from the hedge fund world, we are the first quant manager to offer a concrete strategy based purely on unstructured data.
“At Quoniam we only use data when we are sure that the source is robust and that we understand the data sources.”
Dr Markus Ebner
Head of Multi-Asset
Why don’t you use information from social media?
Currently we work with pure news article providers. This gives us more control of the process. For us at Quoniam, an academic approach is important, which means we only use the data when we are sure that the source is robust and that we understand the data sources. But we will continue to look at data providers to see if they can add value for us. In this respect, we can imagine using data from social media in the next few years.
Will unstructured data significantly influence all investment decisions in the medium to long term?
No, I personally believe that there will always be fund managers who swim against the tide and, for example, bet against a certain market despite supposedly positive sentiment. But I am convinced that, especially in the quantitative area, we can pick up the current market situation much better with such a systematic approach.
How do you see the future of data processing? What role does your house play in this?
We have to stay ahead of the curve. At the moment, Quoniam is definitely a pioneer in this area. And we are working on staying ahead of the curve, because after all, there is information in the data with which, to put it bluntly, you can earn money. Other asset managers recognise this, of course, and will try to tap into the various data sources, so competition will increase. Our challenge is to find new data sources that we can analyse and use to stay ahead and keep the information advantage in unstructured data over other asset managers.