Nowcasting U.S. Economic Activity

The US economy continues to drive global markets, and obtaining reliable and timely updates on its evolution gives investors a competitive edge to navigate uncertainties, and fine-tune their strategies. RavenPackhas published a paper describing the results of a new nowcasting model that uses sentiment analytics to predict economic activity in the US in real time.

Using RavenPack Event Taxonomy and Analytics over a 30-day period provides the best contribution by highlighting specific narratives and recent events. The research has identified that adding RavenPack Analytics reduces average nowcast prediction errors by up to 70%.

The RavenPack nowcasting model (NC-RP) is valuable in terms of timeliness and accuracy. It improves over autoregressive models (AR(1)), nowcasting models based on hard data (NC-B), and US Federal institution benchmarks such as GDPNow, maintained by the Federal Reserve Bank of Atlanta and the Survey of Professional Forecasters of the Philadelphia Fed (SPF).

Watch the video summary