Leverage real-time news sentiment and macroeconomic
variables to track US economic activity daily.
The RavenPack US GDP Nowcasting model leverages proprietary machine
learning models to deliver a prediction of US Economic Activity: The Nowcasting model predicts, on a daily basis, the US GDP growth rate
for the previous quarter Q–1 (Backcast), the current quarter Q (Nowcast),
and the next Q+1, Q+2, Q+3, and Q+4 quarters (Forecasts). Model predictions are presented similar to a PMI index to better capture any sudden changes in the data.
The factors track topics in the news relevant to balance of payments, consumption, employment, taxes, and economic activity, including:
as well as over 30 distinct macroeconomic factors.
The model provides information about which variables altered the final prediction, and the magnitude of its impact. Follow the forecast line
to find out more:
can help quantitative and discretionary investors:
The combination of high-frequency macro factors and traditional variables, with refined sentiment scores
to reduce noise, delivers more accurate forecasts.
The nowcast, backcast, and forecasts aid in identifying economic trends and timing portfolio rebalancing,
especially when combined with inflation and other nowcasts.
The performance of the RavenPack US GDP Nowcasting model can be measured against the Survey of Professional
Forecasters (SPF) in a run-up against the benchmark of an auto-regressive model. The model that produces the RavenPack nowcast improves
on the benchmark by 70% on average, and produces substantially more accurate predictions in the weeks before official figures are released.
Backcasts, nowcasts, and forecasts are the outcome of a multi-step process that includes:
RavenPack analyzes textual content from high-quality sources to identify events that are relevant to the US economy. Documents considered include articles and press releases in English pertaining to specific economic narratives.
Additional filters ensure the data remains focused on contemporary observations, and relevant organizations. In addition, events that relate to the comparison between expectations and actual economic figures are excluded to reduce noise.
For each event identified, RavenPack uses sentiment analysis to compute multiple scores by matching stories usually categorized by financial experts as having a positive or negative financial or economic impact.
Sentiment scores are averaged and smoothed over time then introduced into a dynamic factor model as stationary monthly input variables with daily updates. Several dozen additional macroeconomic factors are also included into the model.
The model produces predictions over 6 quarters, from backcast to forecast, with daily updates.
By providing your personal information and submitting your details, you acknowledge that you have read, understood,
and agreed to our Privacy Statement and you accept our Terms and Conditions. We value your privacy and assure you that we will handle your
personal information with the utmost care and in compliance with our Privacy Statement. You can exercise your rights of access, rectification,
erasure, restriction of processing, data portability, and objection by emailing us at firstname.lastname@example.org.
In accordance with the General Data Protection Rules (GDPR). You also are agreeing to receive occasional updates and communications from RavenPack about resources, events, products, or services that may be of
interest to you.
Please use your business email. If you don't have one, please email us at email@example.com.
Your request has been transmitted. One of our team members will follow up shortly.