Want to keep learning?

This content is taken from the The University of Warwick's online course, Big Data: Measuring And Predicting Human Behaviour. Join the course to learn more.

Skip to 0 minutes and 7 seconds Welcome back. Financial markets are truly fascinating. If you have some idea what they are going to do in the near future, then you have the opportunity to become extremely rich. At the same time, if you are wrong with your forecast, then you can potentially lose a lot of money. From a behavioural point of view, financial markets are truly fascinating because we can study what people have decided to buy or sell. When stock markets became electronic trading platforms in the 1990s, these were, yeah, we can say one of the first big data generators on human behaviour, very detailed in a very detailed fashion, recording what people have decided to do.

Skip to 1 minute and 7 seconds When we started to look at this, we wanted to combine it with all the fascinating other data sources which are out there recording human behaviour. And so we started to incorporate what people are searching for online. Together with Gene Stanley from Boston University and Daniel Reith from the University of Mainz we started to look at searches for company names. One of the very first figures we produced was looking at the number of searches, how many people have searched for the term ‘Lehman Brothers’. If you plot this over time from 2004 until exactly this week, for which Google makes these data available, then you can see a very distinct spike.

Skip to 2 minutes and 9 seconds And this spike coincides with the problems in the financial markets, in particular with the week when Lehman Brothers had to to file for Chapter 11 protection in the United States. If you overlay this with a development of the S&P 500 as one of the major benchmark indices in the world, and you can clearly see that there is a coincidence. So we were really fascinated about the power of Google searches, and we wanted to investigate this relationship further. We downloaded from Google Trends– that’s the service with which Google makes all these data sets available. We downloaded the information on how often people search for company names like Microsoft, Apple, IBM, Intel, and others.

Skip to 3 minutes and 6 seconds And we related this to how many stocks were traded in a given week at the New York Stock exchange. Why did we look at the weekly garularity? That’s very simple. Google provides mainly weekly search volume for all the keywords you could possibly search for. So what we found is that the more often people search for a company in a given week, the more stocks of this particular company are traded at the New York Stock exchange in exactly the same week. There’s a coincidence. And this is a pattern we found holds for all the companies, on average, which form the S&P 500. So this was a study we published in 2010.

Skip to 4 minutes and 2 seconds And this is more or less the starting point of our interest in looking into Google Trends. In one of the next videos, we will show you how you can extend these findings and look at some more interesting patterns, and in particular, also how we can look into price movements rather than volume only. See you then.

Google data and trading in the stock markets

The Internet has now become a central source of information for many. How might searches for information about a company online be related to trading of that company’s stocks?

Watch this video to get a first insight into the relationship between Google data and financial market trading.

Share this video:

This video is from the free online course:

Big Data: Measuring And Predicting Human Behaviour

The University of Warwick

Get a taste of this course

Find out what this course is like by previewing some of the course steps before you join: