Identifying which stocks are peers of each other is of great importance to financial research and applications because it allows for an apples-to-apples comparison of companies and the potential exploitation of information spillover effects. The traditional way of grouping similar firms together is based on standard sector or industry classifications such as GICS or SIC codes. While these are useful as a starting point, they sometimes fail to spot homogenous groups of fundamentally related companies.
For instance, Apple and HP are both in the technology hardware and equipment retailing industry group and are arguably economically linked firms. Apple, on the other hand, is also part of the FAANG collection of stocks, but it falls within a different industry group or even sector compared to its Silicon Valley counterparts.2
Alternative peer groups have been proposed in the academic literature based on firms in the same geographic region,3 customer-supplier data,4 textual similarities in companies’ 10-K business descriptions,5 or comparable technologies based on patent data.6
Making use of analysts’ human intelligence
More recently, Ali and Hirshleifer illustrated the efficacy in linking peers based on shared analyst coverage.7 This approach could be superior for several reasons. Firstly, the shared analyst network makes use of analysts’ human intelligence. Not only is it the analyst’s job to gather all relevant information about fundamentals such as information about related firms, customers, suppliers and competitors; an analyst is also more likely to adapt to changing market conditions than relatively static industry classifications.
Secondly, analyst linkages can uniquely identify connected company pairs rather than linking all firms within a sector, industry or a region. Similarly, analysts can also determine linkages beyond industry and geographic classifications, meaning that a company can be linked to firms both within its own but also outside its sector. Thirdly, the number of shared analysts can serve as a proxy for the strength of the relationship because not all relationships within the same region or sector will be equally important.
Figure 1 | Analyst-based company network for Apple
Source: I/B/E/S, Refinitiv, Robeco. For illustrative purposes only. The figure shows a simplified analyst network of connected firms for Apple as of September 2022. Two firms are connected when the same analyst covers them. The more analysts cover a pair of stocks, the thicker the line between them. The different colors indicate different GICS industry groups.
Figure 1 visualizes the connected company networks based on shared analyst coverage for Apple. The different colors represent different GICS industry groups. Classic information technology firms such as HP (technology hardware and equipment, purple), IBM (software and services, green) and Intel (semiconductors and semiconductor equipment, grey) are connected with Apple via shared analysts. However, we also observe that FAANG stocks from other sectors such as Meta Platforms, Alphabet, Netflix (all media and entertainment, pink) and Amazon (retailing, orange) are also closely related to Apple.
Therefore, we can conclude that the shared analyst coverage approach is an alternative and adaptive way to identify relevant firm peers.
Footnotes
1 Hanauer M. X., and Huisman R., October 2022, “Shared analyst momentum: extracting signals from connected company networks”, Robeco article.
2 Meta Platforms (formerly known as Facebook), Netflix and Alphabet (formerly known as Google) are in the ‘media and entertainment industry group which falls under the communication services‘ sector. Amazon is in the retailing industry group which falls under the consumer discretionary sector.
3 Parsons, C. A., Sabbatucci, R., and Titman, S., October 2020, “Geographic lead-lag effects ”, Review of Financial Studies.
4 Cohen, L., and Frazzini, A., August 2008, “Economic links and predictable returns”, Journal of Finance.
5 Hoberg, G., and Phillips, G., October 2010, “Product market synergies and competition in mergers and acquisitions: a text-based analysis” The Review of Financial Studies; and Hoberg, G., and Phillips, G., October 2016, “Text-based network industries and endogenous product differentiation” Journal of Political Economy.
6 Bloom, N., Schankerman, M., and Van Reenen, J., July 2013. “Identifying technology spillovers and product market rivalry”, Econometrica; and Lee, C., Sun, S. T., Wang, R., and Zhang, R., June 2019, “Technological links and predictable returns”, Journal of Financial Economics.
7 Ali, U., and Hirshleifer, D., June 2020, “Shared analyst coverage: unifying momentum spillover effects”, Journal of Financial Economics.
免責聲明
本文由荷宝海外投资基金管理(上海)有限公司(“荷宝上海”)编制, 本文内容仅供参考, 并不构成荷宝上海对任何人的购买或出售任何产品的建议、专业意见、要约、招揽或邀请。本文不应被视为对购买或出售任何投资产品的推荐或采用任何投资策略的建议。本文中的任何内容不得被视为有关法律、税务或投资方面的咨询, 也不表示任何投资或策略适合您的个人情况, 或以其他方式构成对您个人的推荐。 本文中所包含的信息和/或分析系根据荷宝上海所认为的可信渠道而获得的信息准备而成。荷宝上海不就其准确性、正确性、实用性或完整性作出任何陈述, 也不对因使用本文中的信息和/或分析而造成的损失承担任何责任。荷宝上海或其他任何关联机构及其董事、高级管理人员、员工均不对任何人因其依据本文所含信息而造成的任何直接或间接的损失或损害或任何其他后果承担责任或义务。 本文包含一些有关于未来业务、目标、管理纪律或其他方面的前瞻性陈述与预测, 这些陈述含有假设、风险和不确定性, 且是建立在截止到本文编写之日已有的信息之上。基于此, 我们不能保证这些前瞻性情况都会发生, 实际情况可能会与本文中的陈述具有一定的差别。我们不能保证本文中的统计信息在任何特定条件下都是准确、适当和完整的, 亦不能保证这些统计信息以及据以得出这些信息的假设能够反映荷宝上海可能遇到的市场条件或未来表现。本文中的信息是基于当前的市场情况, 这很有可能因随后的市场事件或其他原因而发生变化, 本文内容可能因此未反映最新情况,荷宝上海不负责更新本文, 或对本文中不准确或遗漏之信息进行纠正。