How CloudQuant predicts macroeconomic factors early using Prosper and AWS Data Exchange
Hedge funds and equity investors are continuously seeking to create alpha in their portfolios to improve investor returns and minimize risk. Alpha (α) is a term used in investing to describe an investment strategy’s ability to outperform overall market, or its “edge.” To meet this need, “Alternative Data,” data not gathered from traditional sources, is in demand to power new analytics that forecast trends for industries and stocks and improve returns. Investors who have access to greater information have a greater advantage over other investors.
CloudQuant seeks to predict macroeconomic factors three weeks early
CloudQuant turns alternative data into solutions for generating alpha and improving portfolio profits through its artificial Intelligence and Alternative Data platform. CloudQuant is a member of the AWS Partner Network (APN) and uses Prosper Insights & Analytics and AWS Data Exchange to create predictive analytics.
For decades, investors have relied on government data, in various degrees, for understanding the direction the economy is moving. A growing economy may be a signal for stock appreciation, and a declining economy may be a sign of falling stock prices. The key is to identify in advance economic direction through macroeconomic indicators.
CloudQuant tasked its AI researchers to predict global macroeconomic factors three to four weeks ahead of the actual government release dates. The company has connected over 13,000 alternative datasets to its alternative data fabric and analytics tech stack.
Using Prosper data to make accurate early predictions
Over the past 20 years, Prosper Insights & Analytics has been conducting the largest scientific and representative monthly survey of consumer behaviors, motivations, and future purchase intentions. Prosper’s data consists of consumer responses to direct questions and requires no additional assumptions. Many other alternative data sets are derived from clicks or swipes, requiring numerous unverifiable assumptions regarding the true identity, characteristics, and purpose for the transaction. In comparison, Prosper data is more factual, since it requires no unverifiable assumptions. The data delivers timely information on changing patterns of major categories of consumer expenditures, where the consumer is shopping, including online to chain stores, and monitors macroeconomic performance and financial conditions. This data offers intelligence on both current and planned behaviors in the next 90 days to six months.
Prosper Insights & Analytics is a verified Independent Software Vendor (ISV) in AWS Marketplace as well as a trusted partner and resource for the National Retail Federation (NRF), which has used Prosper data on consumer holiday spending and trends since 2003.
With the Prosper data and other publicly available data sources, CloudQuant’s research team was able to predict a variety of important economic factors with a useful degree of accuracy prior to government release of data.
CloudQuant enables meaningful predictions 24-48 days before official data release
CloudQuant used Prosper data to refine its predictive models to more accurately predict the Consumer Price Index (including CPI sub-components), Non-Farm Payroll, Housing Starts, Average Hourly Earnings, Advance Retail Sales, and Person Consumption. It was able to make these predictions 24-48 days before the official data was released by the government. Analytics like these are called macroeconomic signals. They are a key way for traders to gain insight into the state of the economy, which can have a significant influence on the direction of stock market movements. Most importantly, the advanced delivery of the analytics allows traders to make decisions before government data is released which can improve alpha.
CloudQuant presented the results of this analysis in June of 2021 at the JP Morgan Global Institutional Investor Conference. It demonstrated an ability to predict the directional change in these economic factors of 63.1% directional accuracy for CPI with an Information Coefficient (IC) of 0.31. Mean directional accuracy (MDA), also known as mean direction accuracy, is a measure of prediction accuracy of a forecasting method in statistics. It compares the forecast direction (upward or downward) to the actual realized direction. The Information Coefficient shows how closely the analyst’s financial forecasts match actual financial results. Prosper’s data helped to predict the change in another factor, Average Hourly Earnings, with an 84.6% (IC 0.17) directional accuracy and Advanced Retail Sales changes with 64.3% directional accuracy (IC 0.13). Both the directional accuracy and information coefficient measures demonstrate an advantageous input for the predictive analytics
“Using Prosper’s data to supercharge predictive analytics is a perfect example of working smarter rather than working harder. Sourcing the right alternative data from AWS Data Exchange and simplifying integration to existing processes with CloudQuant Liberator™ data fabric is the key differentiator for success,” said CloudQuant Founder and CEO, Morgan Slade.
To find out more about Prosper datasets, visit the AWS Marketplace Data Exchange to access relevant consumer intent data to create accurate predictive analytics, better forecast markets, or enhance existing internal data. Prosper US Signals datasets include:
US Signals – Macro Economic and Consumer Purchase Intentions
US Signals – Retail Economy and Consumer Spending Forecast
US Signals – Retail Softlines Consumer Spending Forecast
US Signals – Retail Hardlines Consumer Spending Forecast
US Signals – Grocery, Health & Beauty Consumer Spending Forecast
The content and opinions in this post are those of the third-party author, and AWS is not responsible for the content or accuracy of this post.
About the author
Gary Drenik is President/CEO and co-founder of Prosper Business Development. With a 30-year tradition of innovation, Prosper has provided market leadership and developed contemporary solutions to help companies navigate change. Recognized multiple times as an Inc 5000 company, Prosper brands include Prosper Insights & Analytics™, Prosper Technologies™ and ProsperChina™.
Drenik has been a Forbes contributor since 2008 focusing on innovation and data analytics. He is a co-author of “A Critical Look at Media Planning” published by the Journal of Consumer Behaviour and is frequently quoted by the press.
Drenik holds a B.A. from Baldwin-Wallace College. He lives in Columbus, Ohio.