Hearing “SURPRISE!” can be lots of fun at a birthday party, but not so much for investors and traders trying to judge the market-moving potential of economic indicators and FX rates. Let StarMine SmartEconomics light the candles and illuminate sharper insight.
StarMine has applied its proprietary SmartEstimates® methodology to forecasts of macroeconomic data and FX rates to create a SmartEstimate that is more accurate than the simple consensus forecast. StarMine SmartEconomics marries the breadth of Thomson Reuters Datastream economic data with industry-leading Reuters polling data to rigorously assess the historical accuracy of each contributor at every point in time on every indicator for which the contributor had a forecast.
The indicator-specific StarMine historical accuracy score for each forecaster then determines the weight that each forecast receives in the SmartEstimate. Backtests show that the SmartEstimate correctly predicts the direction of macro and FX surprises relative to the consensus forecast about 61% of the time when the SmartEstimate is significantly different from the consensus. For some important indicators, the hit rate has been even better.
A recent example
Based on our research, as well as a large body of academic research, the U.S. non-farm payrolls release is the single most market-moving economic indicator. The non-farm payrolls number represents the number of jobs added or removed from the private sector of the U.S. economy since the previous month, not including farm workers. The equity, FX, fixed income, and commodities markets all react strongly to the payrolls releases. The non-farm payrolls number is released every month, normally on the first Friday of the month at 8:30 a.m. Eastern Time by the U.S. Department of Labor as part of a comprehensive labor market report.
On Thursday, March 6, markets were bracing for a weak payrolls report the next day, which we had reported in our own Reuters Insider:
Better analysis
However, those who checked the StarMine SmartEstimate forecast for U.S. non-farm payrolls (RIC = USNFAR=ECI) would have seen StarMine was actually predicting a positive surprise.
The SmartEstimate stood at 150,900 jobs, while the consensus was for only 149,000 jobs, yielding a Predicted Surprise (SmartEstimate – Consensus) of +1,900 jobs. While this difference may seem small, over the last two years it has been “directionally correct” (correctly predicted the sign of the actual surprise) 72% of the time.
When the payrolls report was released on 7 March, the actual value came in at 175,000 jobs, an actual surprise (Actual – Consensus) of +26,000. The reaction in the FX market was immediate and strong. Below we show the USD/JPY spot rate at 1-minute intervals around the release. The big spike upwards occurred at the time of the release (13:30 GMT).
More key data
One example doesn’t prove performance, of course, but looking at many examples can make a very strong case.
The plots below display the forecasts, consensus, SmartEstimate and actual released values for U.S. non-farm payrolls over the last two years. In the top panel we show the forecasts and in the bottom panel we show the Predicted Surprise. If the sign of the Predicted Surprise matched that of the actual surprise, we call that a “Hit” and circle those Predicted Surprises in green. In 18 of the last 25 payrolls releases (72%) the SmartEstimate has correctly predicted the direction of the actual surprise. Since the SmartEconomics model has been running live (since mid-June 2013) the hit rate has been 8 out of 11 (73%).
Follow the SmartEstimate
Is the SmartEstimate always right? No. But as we have seen, it tends to be more right than not and can help give investors a leg up in predicting the direction of economic surprises for some of the most important market-moving economic releases.
StarMine SmartEconomics is available in Thomson Reuters Eikon and as a data feed. For more information on StarMine SmartEconomics and how you can gain an edge in macroeconomic and FX forecasts, watch our recent webinar or contact your Thomson Reuters representative or StarMine.QuantConsulting@ThomsonReuters.com
Receive stories like this to your inbox as they are published. Subscribe here and follow us @Alpha_Now on Twitter. If you are looking to access Thomson Reuters data or analytics, register for a free trial.
