In recent years, there has been a rapid increase in the amount of data being generated from a wide variety of sources, both by individuals and by companies. Traditionally, the most important datasets for traders have consisted of data describing price moves. For macro traders, economic data has also been a key part of the trading process. However, by augmenting their existing processes with these new alternative datasets, traders can gain greater insights into the market. In this chapter, we delve into the topic of alternative data and big data. We split our discussion into three parts.
In the first section, we seek to define general concepts around big data and alternative data. We explain why data is being generated at a rapidly increasing rate and the concept of ‘exhaust data’. We discuss various approaches to developing models to describe the market, comparing traditional approaches to machine learning. We elaborate on the various forms of machine learning and how they might be applied in a financial setting.
In the next section, we focus more on general applications for alternative data in macro trading. We note how it can be used to improve economic forecasts, for example, or in the construction of nowcasts. Real‐life examples of big data and alternative datasets such as those derived from newswires and social media are also listed.
In the final section, we go into more ...