5 questions on our new DataEdge alternatives strategy

In this Q&A, Laurent Joué and Marc Pellaud, Co-Leads of the LOIM DataEdge alternatives strategy, explore how actionable insights from big data can help better predict company earnings. How do alternative data sets improve our understanding of consumer behaviour and its impact on company results in order to uncover alpha opportunities?

Need to know: Our DataEdge alternatives strategy uses big data to identify trends in consumer behaviour in real time. Such data can be used to improve the forecasts of company earnings Within the DataEdge investment universe, our research has found that earnings surprises and equity returns are tightly connected – both on the upside and downside – and can generate alpha opportunities LOIM’s investment approach combines innovative alternative data research with systematic investment expertise. We then apply this ‘information edge’ to stock selection

 

1. How does LOIM’s DataEdge strategy use alternative data to improve the forecasting of key company indicators?

The strategy relies on the use, acquisition and analysis of alternative data points across multiple geographies, sectors and sources. These include:

• credit card transactions
• digital receipts
• web traffic
• activity on apps
• geolocation services

Thanks to new advances in data science, it is now possible to structure and filter this data to identify consumer behaviour trends in real time. Systematic and quantitative analysis methods are then applied to predict how patterns could impact the key performance indicators (KPI) of listed companies, such as sales, revenues, users, new customers and engagement patterns.

To generate alpha potential, the strategy aims to identify earnings surprises, as well as track inflection points in the trends of KPIs over several quarters.

 

2. What is the strategy’s investment universe?

This alternatives strategy is global and sector neutral. The investment universe is currently US centric, and focuses on consumer, technology, industrial and healthcare companies. In term of market capitalisation, it is predominantly exposed to small and mid-caps.

This universe is shaped by the availability and quality of the alternative data we seek, and is expected to evolve and grow over time. The current¹ eligible universe covers 640 companies listed on US markets, and the strategy monitors 200-250 companies day by day. Figures 1 and 2 provide breakdowns of the universe by sector and market cap.

FIG 1. Sectors comprising the DataEdge universe

Source: LOIM. For illustrative purposes only. As at March 2024. Allocation subject to change.

 

FIG 2. Market capitalisation breakdown of the DataEdge universe

Source: LOIM. For illustrative purposes only. As at March 2024.

 

3. What role do earnings surprises play? Are they highly linked to equity returns for the DataEdge universe?

Our research² shows that within the DataEdge universe, earnings surprises and equity returns are tightly connected, both on the upside and downside. Using this universe, we assessed the relationship between earnings surprises and equity returns by looking at three factors.

• Relevance. We used metrics such as correlation and the probability distribution of gains³, as well as regression analysis⁴. Performing a regression of earnings surprises versus equity returns on earnings released over the last 20 years and five years, our model found a convincing and valid relationship on the DataEdge universe. This indicates that earnings surprises can drive a significant part of price movements⁵.
  
  Our study also showed potentially attractive metrics for the past five years in term of relevance. The correlation was stable, even slightly higher, as well as the average probability of gains, which remained close to historical levels.
   
• Market reward. This refers to equity returns on average for positive and negative earnings surprises. Over the 20-year period, we found that the market has rewarded positive earnings surprises by +1.9% on average and detracted -3.6% on average for a negative earnings surprise. Further, 86% of the universe had an average positive return for observations of positive earnings surprises and 80% of the universe had an average negative return for observations of negative earnings surprises. The last five years mirrored these results: the market has rewarded positive earnings surprise by +1.7% on average, and -4.1% on average for negative earnings surprise.
   

FIG 3. Market reward relative to positive and negative earnings surprises

Source: LOIM. For illustrative purposes only. Past performance is not an indicator of future results. 20 years covers January 2004 to January 2024 and 5 years covers 2019-2024.

 

• The surprise bias. This analysis measured the imbalanced proportions of stocks that beat or miss the consensus forecast. The error of consensus versus the reported value was measured by the mean absolute percentage error (MAPE). Over the 20-year period, the average percentage of positive earnings surprises was 72% vs 28% for negatives surprises. Meanwhile, 88% of the universe has exhibited positive earnings surprises more than 50% of the time. The average MAPE across the universe was 4.14%.
  
  We noted no significant changes in the surprise bias over the last five years compared to the 20-year period. The findings suggest that consensus expectations tend to consistently underestimate corporate earnings in the US. On the other hand, we found a significant increase in the consensus error over recent rolling periods, which appears to be explained by the Covid-19 pandemic.

 

FIG 4. Surprise bias

Source: LOIM. For illustrative purposes only. Past performance is not an indicator of future results. 20 years covers 2004-2024 and 5 years covers 2019-2024.

 

4. Do alpha opportunities arise from the strong link between earnings surprises and stock performance? 

The relationship between earnings surprises and stock performance has been examined for more than 50 years in the academic literature⁶, and our research confirms the relationship is valid for the DataEdge universe. As the flow of information accelerates and becomes more widely available, and the use of alternative data grows significantly, questions arise about the alpha potential that can be extracted from earnings surprises.

We formulated a theoretical strategy⁷ to gauge the alpha potential of surprise earnings on the DataEdge universe. In short, the strategy takes a long position on positive earning surprises and the reverse for the negative ones, based on historical price moves. We refer to this strategy as ‘perfect foresight’ as it is not implementable using information that was public at the time of portfolio construction. However, even though the theoretical strategy cannot be traded, it nonetheless gives us an understanding of alpha potential, its behaviour over time, as well as the magnitude of price responses to new accounting information.

The results of our analysis showed that significant alpha potential is available for strategies able to predict earnings revisions. The information ratio⁸ after trading costs ranged from 3.5 and more than 4 depending on the period, and was stable over shorter or longer periods.

Based on these results and on our average forecast’s hit ratio on earnings revisions⁹, which is close to 70%, we would expect the strategy to have an information ratio of 1.6, according to parameters set by Grinold and Kahn research¹⁰

This reaffirms that earnings surprises can drive significant price movements and present abundant alpha opportunities, in our view.

 

5. What is LOIM’s expertise in systematic alternatives and what framework do you adopt?

For the DataEdge strategy, we combine advanced alternative data research with systematic investment expertise. As Lead Portfolio Managers, we have a combined 36 years of experience in systematic alternatives strategies. We work with LOIM’s quantitative investment team, which is composed of 21 investment professionals and manages more than USD 10 billion¹¹. We also partner with a pioneering data aggregator with close links to our 1798 Alternatives franchise.

Our differentiated investment framework is not a simple mix of fundamental and quantitative approaches. Instead, we aim to isolate pure idiosyncratic alpha by applying our information edge to stock picking. Ultimately, this creates a contrarian and diversified portfolio that invests in stocks where our models predict large potential divergence from consensus market expectations.

For professional investors use only

This document is a Corporate Communication and is intended for Professional Investors only. 

This document is a Corporate Communication for Professional Investors only and is not a marketing communication related to a fund, an investment product or investment services in your country. This document is not intended to provide investment, tax, accounting, professional or legal advice.

This document is issued by : 

Lombard Odier Asset Management (Europe) Limited (hereinafter the “Company”). The Company is authorised and regulated by the Financial Conduct Authority (the “FCA”), entered on the FCA register with registration number 515393. 

This document is approved at the date of the publishing. The Company is clustered within the Lombard Odier Investment Management Division (“LOIM”) of Lombard Odier Group which support in the preparation of this document and LOIM is a trade name.

Any opinions or forecasts provided are as of the date specified, may change without notice, do not predict future results and do not constitute a recommendation or offer of any investment product or investment services.

This document is the property of LOIM, is provided for information purposes only and is addressed for the recipient exclusively for its personal use. It may not be reproduced (in whole or in part), transmitted, modified, or used for any other purpose without the prior written permission of LOIM. It is not intended for distribution, publication, or used for any other purpose without the prior written permission of LOIM. 

The contents of this document are intended for persons who are professionals and who have been vetted by LOIM and assessed as suitable to the investment matters set out in this document and in respect of whom LOIM has received an assurance that they are capable of making their own investment decisions and understanding the risks involved in making investments of the type included in this document or other persons that LOIM has expressly confirmed as being appropriate recipients of this document. If you are not a person falling within the above categories, you are kindly asked to either return this document to LOIM or to destroy it and are expressly warned that you must not rely upon its contents or have regard to any of the matters set out in this document in relation to investment matters and must not transmit this document to any other person. This document contains the opinions of LOIM, as at the date of issue or completeness of the information contained in this document, nor does it accept any liability for any loss or damage resulting from its use. All information and opinions as well as the prices indicated may change without notice.

The contents of this document has not been reviewed by any regulatory authority in any jurisdictions and does not constitute an offer or a recommendation to subscribe for any securities or other financial instruments or products.   

It contains opinions of LOIM, as at the date of issue. These opinions and information contained herein in this document does not take into account all the specific circumstances of the addressee. Therefore, no representation is made that the information presented in this document are suitable or appropriate to the individual circumstances of any investors. Tax treatment depends on the individual circumstance of the investor and may be subject to change in the future. LOIM does not provide tax advice. 

The information and analysis contained herein are based on sources believed to be reliable. While LOIM uses its best efforts to ensure that the content is created in good faith and with greatest care, it  does not guarantee the timeliness, accuracy, validity, reliability or completeness of the information contained in this document, neither does it warrant that the information is free from errors and omission not does it accept any liability for any loss or damage resulting from its use. All information and opinions as well as the prices indicated may change without notice. Particular contents of third parties are marked as such. LOIM assumes no liability for any indirect, incidental or consequential damages that are caused by or in connection with the use of such content. 

The Source of the data has been mentioned wherever it was available. Unless otherwise stated, the data is prepared by LOIM. 

Not for US Person: This corporate communication is not intended for any "U.S. Person" as defined in Regulation S of the Act, as amended or pursuant to the 1940 United States Investment Company Act as amended and will not be registered pursuant to the 1940 United States Investment Company Act as amended, or pursuant to other US federal laws. Neither this document nor any copy thereof may be sent, taken into, or distributed in the United States of America, any of its territories or possessions or areas subject to its jurisdiction, or to or for the benefit of a United States Person. For this purpose, the term "United States Person" shall mean any citizen, national or resident of the United States of America, partnership organized or existing in any state, territory or possession of the United States of America, a corporation organized under the laws of the United States or of any state, territory or possession thereof, or any estate or trust that is subject to United States Federal income tax regardless of the source of its income.
 
Data Protection: You may be receiving this Communication because you have provided us your contact details. If this is the case, note that we may process your personal data for direct marketing purposes. For more information on Lombard Odier’s data protection policy, please refer to www.lombardodier.com/privacy-policy 
 
©2024 Lombard Odier IM. All rights reserved.