The StarMine research team scrutinizes earnings forecasts toward the end of each quarter, in search of companies that its StarMine SmartEstimate® and Predicted Surprise models signal are most likely to beat – or fall short of – analysts’ consensus forecasts. Of the ten companies that we believed had a high probability of recording an earnings surprise, we used these models to accurately predict the direction of the subsequent earnings surprise in seven of those companies in the latest quarterly reporting period for North American companies. That gives our forecasting process a 70% accuracy rate.
The Predicted Surprise percent relies on both StarMine SmartEstimate and the I/B/E/S consensus estimate, and is a measure of the percentage difference between the two. Our research shows that when the magnitude of the Predicted Surprise is significant this is a good predictor of the direction of future earnings revisions. Historically, the Predicted Surprise accurately forecasts the direction of subsequent surprises correctly 70% of the time. History also shows that being able to predict accurately the direction of future earnings revisions and earnings surprises gives investors a good chance of predicting the direction of changes in the stock price. That’s because a company’s stock price typically responds to a piece of unexpectedly good earnings news by rising, and vice versa. Indeed, that correlation between stock price movements and earnings news has become so well understood that it’s no longer profitable for investors to wait until after the surprise is reported. Instead, the investor has to anticipate those surprises.
The StarMine SmartEstimate, upon which this earning surprise forecast relies, is designed to be a more accurate version of analyst consensus data. It’s a weighted reformulation of analyst estimates, designed to place more emphasis on the most recent forecasts from the analysts with the best track records.
The chart below shows the results of our team’s study of ten North American companies that had a high probability of reporting an earnings surprise when they reported their results for the fourth quarter of 2011.
*Although the team’s prediction for the direction of the Amazon.Com surprise wrong for earnings per share (EPS), it is noteworthy that the company stock was punished for missing on the revenue consensus and guided to lower revenue targets for the following quarter.
Learn more about how StarMine analytics can help you pinpoint critical developments in your portfolio or watch list. Request a free trial today.
