Skip to content
Thomson Reuters
Investment Management

A star strategy for predicting earnings season hits and misses

David Aurelio

18 Jan 2018

A starfish. Photography: Natasha Elkington
Photography: Natasha Elkington

It’s January and the beginning, once again, of a busy period for financial market participants as earnings season anticipation begins.

Earnings surprises and consensus revisions are well-known drivers of stock price movements and StarMine has a proven ability to predict these surprises and revisions, via SmartEstimate®.

This model is more accurate than the consensus, as it puts more weight on recent forecasts from top-rated analysts.

StarMine provides a suite of proprietary alpha-generating analytics and models spanning sectors, regions, and markets

Watch video — The evolution of Lipper from its inception to the present day

Every quarter, the Thomson Reuters StarMine team reveals 10 companies with a high probability of reporting quarterly earnings surprises in the direction that is called: five likely to beat the analysts’ consensus estimate and five likely to fall short.

Earnings season – Monitor developing trends as results unfold. Find out more.

When SmartEstimates® diverge significantly from Consensus, you can anticipate the occurrence of earnings surprises with an accuracy rate of 70 percent.

Piles of dollar bill notes

Revenue SmartEstimates® are even more predictive of surprises, with a historical accuracy rate of 78 percent

Let’s take a look at the analysis.

Companies picked for earning hits

The StarMine team selected five North American companies, using Thomson Reuters’ Eikon Screener, that are expected to beat 2017 Q4 earnings estimates, based on SmartEstimate® and Predicted Surprise data.

The North American selections for 2017 Q3 were 100 percent accurate. Historically (since 2011 Q4), the picks have demonstrated an accuracy rate of about 80.8 percent giving investors an edge ahead of earnings announcements.

Caterpillar's earnings as seen on Thomson Reuters StarMine
Caterpillar’s earnings as seen on Thomson Reuters StarMine

The North American picks for positive Predicted Surprises in 2017 Q4 are:

Caterpillar Inc (CAT.N); T-Mobile US Inc (TMUS.O); Kosmos Energy Ltd (KOS.N); Activision Blizzard Inc (ATVI.O); and Carrizo Oil & Gas Inc (CRZO.O).

Thomson Reuters StarMine
Thomson Reuters StarMine

Keep up-to-date with all the most current articles and reports produced by the research team with Lipper Alpha Insight Newsletters

Companies picked for earnings misses

The StarMine team also selected five North American companies that are expected to miss 2017 Q4 earnings estimates.

Get access to trusted news, data, and analytics, all filtered by relevance to your exact needs with Thomson Reuters Eikon

The North American selections for 2017 Q3 misses were 41.2 percentage points above the S&P 500’s 17 Q3 earnings miss rate.

Expedia's earnings as seen on Thomson Reuters StarMine
Expedia’s earnings as seen on Thomson Reuters StarMine

Historically (since 2011 Q4), the negative Predicted Surprise picks have demonstrated an accuracy rate of about 64.4 percent, giving investors an edge ahead of earnings announcements.

The North American picks for negative Predicted Surprises in 2017 Q4 are:

PG&E Corp (PCG.N); Expedia Inc (EXPE.O); Discovery Communications Inc (DISCK.O); TransAlta Corp (TA.TO); GTT Communications Inc (GTT.N).

Learn more about Thomson Reuters StarMine Quantitative Analytics and how you can navigate through earnings season.

Thomson Reuters StarMine

Adjusting for the optimism of the sell-side analyst Will equity capital markets hold their strength? The platform to digital transformation in financial trading Global capital markets have a date with reality MiFID II progress summary report: Why there’s still work to do Quantamental investing: Why AI still needs the human touch Connecting apps with Eikon Side by Side Integration API Capital markets off to a flying start in 2018 Winners and market views from U.S. Thomson Reuters Lipper Fund Awards Powering asset management – Overcoming research unbundling