Distilling the Macroeconomic news flow
As macroeconomic news and financial data is reported with ever increasing frequency, this research proposes a simple but effective method for extracting information relevant for the tracking of current economic conditions.
Traditionally the state of the economy is assessed with low-frequency observations of economic aggregates using data collected over weeks, months and even quarters. One example of this in the UK is the advance estimate of GDP, released about a month after every quarter. With modern information sources however, every day market participants are exposed to a constant stream of macroeconomic news, providing near immediate information that can have significant effects on financial markets.
The purpose of this research therefore is to propose a simple cross-sectional technique to extract daily factors from economic news released at different times and frequencies. This distills news flow into a small set of indicators describing four aspects of the economy:
- macroeconomic sentiment.
This approach can effectively handle the large number of announcements relevant to the tracking of current economic conditions, therefore providing more timely and accurate forecasts of future changes in economic conditions than other real-time forecasting methods. The approach allows the data to speak as much as possible, and the results demonstrate that a simple and unstructured method can still deliver a timely measurement of the state of the economy.
A draft version of the research paper can be downloaded at the link below. It comprises data, methodology, results and conclusions. The full version is forthcoming in the Journal of Financial Economics.