In last week’s post titled “Benefiting from the SmartEstimate during times of uncertainty”, we highlighted Zalando (ZALG.DE) as a company that had a strong positive StarMine Predicted Surprise for both FY2020 and upcoming 20Q2 earnings release.
Zalando reported 20Q2 earnings on August 11 and beat earnings expectations, posting an actual result of €0.47 vs consensus of €0.45, resulting in a 5.62% surprise factor. As a reminder, the SmartEstimate can be seen as a leading indicator that is predictive of future analyst revisions.
For customers who wish to look for companies with positive analyst sentiment, we can use the Eikon screener to screen for companies with a positive Analyst Revisions Model (ARM) score. The ARM model is a stock ranking model that is designed to predict future changes in analyst sentiment by looking at changes in estimates across EPS, EBITDA, and revenue over multiple time periods. The model also incorporates the Predicted Surprise which can be viewed as a complimentary signal (i.e. strong analyst revisions combined with a positive Predicted Surprise creates a more robust signal).
Exhibit 1 highlights Zalando having the strongest ARM score of all companies within the STOXX 600, with a score of 100. In this example, we are using the ARM Region Rank, which will rank a company based on its peers in a region (i.e. Developed Europe).
Exhibit 1: ARM Score for STOXX 600 Constituents
Source: Refinitiv Eikon
In Exhibit 1, we also display “ARM Region Rank Change (D30)” which shows how a company’s ARM score has changed over the last 30 days. This field is also available for screening purposes which can help identify companies who have had a large change in ARM score over a short time-period, which can be viewed as a proxy for financial momentum. CNH Industrial NV (CNHI.MI) is a good example, who has seen its ARM score increase from 6 to 97 (change of 91) over the last 30 days.
Continuing our focus on Zalando, Exhibit 2 highlights the ARM model as seen in Refinitiv Eikon. As explained previously, it has an ARM score of 100, which is a percentile-based rank comparing Zalando to its regional peers in Developed Europe. We can see the components that go into the ARM model which include Revenue, EBITDA, EPS, and Recommendations and the associated score within each component.
StarMine also allows for the ARM model to rank companies on both a Global and Country-level basis for additional granularity.
Exhibit 2: Zalando ARM Model
Source: Refinitiv Eikon
The ARM model also displays numerous graphs which shows how the consensus and SmartEstimate has changed over time. In Exhibit 3, we observe how estimates have moved for Revenue, EBITDA, and EPS over the upcoming quarter, current fiscal year, and next fiscal year – providing a comprehensive outlook on analyst sentiment.
In this example, we can see how estimates across the board are moving in an upward trajectory, which is caused by positive analyst revisions. It is also encouraging to see the SmartEstimate (gold line) be higher than the consensus (blue line). We can confirm our prior statement that the SmartEstimate can be seen as a leading indicator and predictive of future analyst revisions, as the consensus is consistently playing catch-up to the SmartEstimate. As the consensus is increasing, investors can benefit the behavior of analysts revising estimates upward can potentially have a material impact on the stock price.
Exhibit 3: Zalando ARM Model
Source: Refinitiv Eikon
The ARM model also shows both the Blended Mean Change and Predicted Surprise for each graph. The Predicted Surprise was explained in last week’s post but is also shown here as a complimentary signal – if a company is seeing analysts increase their estimates in addition to having a positive Predicted Surprise, this can be viewed as a more robust signal.
The final component of the ARM model is shown in Exhibit 4, which is the ‘Recommendations’ component. Refinitiv Eikon has a 1-5 ranking system, 1 being a strong buy and 5 being a strong sell. Looking at the current recommendation score of a company on its own is not entirely meaningful. Instead, it is more valuable to see how the recommendation score has changed over time, which is what the ARM model uses. The mean recommendation change for Zalando is -0.10, which is a blended score and indicates that the mean recommendation is moving upwards (1 = strong buy). For example, Zalando had a mean recommendation score of 2.681 two months ago compared to a current score of 2.471, indicating that analysts are increasing their recommendation rating on the company.
Exhibit 4: Zalando ARM Model
Source: Refinitiv Eikon
To wrap up, the ARM model can be helpful to investors looking for a comprehensive signal on analyst sentiment that can be predictive of future revisions. To view a full list of models offered by StarMine, please click here.
