The Book of Alternative Data

CHAPTER 19Market Data

19.1. INTRODUCTION

It might be unusual to include market data as “alternative.” After all, if we consider market data such as daily closing prices, these are widely available to investors, in many cases for free from websites like Google Finance and Yahoo Finance. However, if we delve more deeply, we note that there is a wide discrepancy between the various asset classes, the frequency of data, and also the granularity. If we would like very high-frequency market data and in particular data that gives us an idea of quotes for different sizes (in other words market depth), this data is much more expensive. Furthermore, the sheer size of such datasets makes them more difficult to consume.

For very illiquid asset classes, which trade very infrequently, simply getting a daily time series of prices might be close to impossible. Even for very liquid asset classes, the breadth of information can vary. In equities, data such as volume traded is very common. This contrasts with FX, which is predominantly an OTC market and comprehensive volume data is more difficult to source.

In this chapter, we will talk about two examples of alternative datasets derived from market data in FX. First, we talk about aggregated FX flow data collected by CLS and show how it can be used to create systematic trading rules to trade FX. Later, we use high-frequency FX tick data to build up a picture of how liquidity has changed over the years and how it changes by time of day.

19.2. ...