An independent review of RIWI’s proprietary datasets on US employment trends by quantitative hedge fund researchers Radu Ciobanu, PhD, and Ernest Chan, PhD has shown that RIWI data outperformed all other benchmarks in predictive accuracy for a surprise in nonfarm payrolls. A “nonfarm payrolls surprise” is when official US monthly nonfarm payrolls data diverge from the consensus prediction of Wall Street economists.
The paper, entitled, “US nonfarm employment prediction using RIWI Corp. alternative data”, showed:
- After weighting the data and performing seasonal adjustment, the application of monthly averages of daily, continuous RIWI sentiment data from December 2013 – October 2017 outperformed all other benchmarks available in predictive accuracy for the “sign” (i.e., positive or negative) of any surprises in official nonfarm payrolls data;
- After cross-validation tests, the average predictive accuracy of “the RIWI Score” for the sign of nonfarm payrolls surprises was 63%; and
- When predicting both the magnitude and the sign of a nonfarm payrolls surprise, combining the RIWI score with public data was critical to enhance predictive insights.
Financial markets mainly react to the surprise and traders who have an information advantage on whether or not to expect a nonfarm payrolls surprise can invest with much more confidence on trades.
The paper also refers to the potential of using RIWI data to more accurately gauge other economic indicators:
“There is great potential for using the RIWI score for predicting the all important Nonfarm Payroll number. But beyond predicting NFP surprises, RIWI’s data have the potential to be a more accurate gauge of the actual U.S. employment situation, and therefore economic growth, than the NFP number.”
A summary of the independent analysis of RIWI’s predictive power for the signs and magnitude of nonfarm payroll surprises can be read here.
Featured image: Shutterstock
