CFA Article Points to Key Alpha Generating Features of the EDS Analytics Platform

November 5, 2017

EDS FirstLook

November 2017

The EDS Comp sheet helps investors identify the right peers using a combination of industry (GICS) classification and Profitability, Growth and Quality factors.

A recent CFA Journal publication, entitled 
Stick to the Fundamentals and Discover Your Peers, written by Jens Overgaard Knudson, Simon Kold and Thomas Plenborg highlights the importance of using this combination approach to "discover" your peers. The end result is a more accurate valuation estimate, leading to higher conviction in price targets and risk/reward calculations, which leads directly to a higher quality portfolio and increased alpha.

The study highlights:

 

  • Identifying comparable companies on the bases of profitability, growth and risk is an essential alternative to the industry classification method.

  • Most investors only use industry classification as a proxy for company comparability.

  • The most accurate results are obtained when both methods are used, as they each capture different fundamental aspects. 

 

EDS takes this concept and puts it into practice in an efficient and scalable way for investors, across over 12,000 equities, in seconds - making our "Comp Sheet" an alpha generating tool for investors. 

 

Click below to see The EDS Analytics Platform in action:

 

  • Limited Brands (LB) - In a few clicks, deep scenario and risk analysis is achievable with EDS - put your ideas into context, discover the right comps, and make sure your price targets are as accurate as possible.

  • Restaurants (CMG) - One click to access the factors that drive any Industry! EDS does all the hard work, running the math in the background and presenting the results in a rich and insightful way. 

 

Excerpts from the Article

The "multiples approach" relies on the theory that perfect substitutes should sell at the same price. This theory implies that the selection of comparable companies is a crucial step. 

 

Most analysts and investors use industry classification as a proxy for company comparability. The use of industry relies on the assumption that companies in the same industry share the same economic characteristics—that is, profitability, risk, and growth. However, this is rarely the case and thus, they should not necessarily trade at the same multiple.

 

Prior studies have demonstrated that identifying comparable companies on the basis of different proxies for profitability, growth, and risk may be a useful alternative to the industry classification method. In our study, we defined a target company’s peer group as the companies with the smallest sum of absolute rank differences across the target company’s variables of interest. The SARD approach is similar to the clustering algorithm known as the “Manhattan distance.”  Results show that the SARD (sum of absolute rank differences) approach yields significantly more accurate valuation estimates than the industry approach. 

 

In addition, a combination of the SARD approach and the industry classification approach results in even more accurate valuation estimates, suggesting that the two approaches capture different aspects of peers’ fundamentals. Our findings are robust across time, company size, and varying numbers of peers, which, coupled with the practical applicability of the approach, makes it a highly relevant tool in a financial analyst’s toolbox.

 

EDS Approach

One of the core differentiators of the EDS platform is the ability to easily build the relevant list of comparable companies using a combination of the traditional industry classification approach (GICS) and similar fundamental measures,  such as Profitability, Growth and Quality.

 

Exhibit 1. EDS Dynamic Comp Sheet (LB): Limited Brands  - 18 companies within the US consumer discretionary sector with similar Profit and Growth levels highlight potential undervalution for LB. Comparable companies on a profitability and growth basis trade at an EV/EBITDA multiple of over 10X vs. LB at 7.2X. 

 

Dynamic Regressions and a Warranted Multiple
Bhojraj and Lee (2002) estimated a series of annual cross-sectional regressions of EV/sales (enterprise value divided by sales) and P/B on eight different proxies for profitability, growth, and . They referred to this prediction as a company’s “warranted multiple,” which is used to identify comparable companies. Bhojraj and Lee found that their warranted multiple provides more accurate valuation estimates than alternative selection methodologies, such as industry classification and size.

 

EDS Approach

EDS's Valuation Regression allows fundamental investors to choose relevant fundamental measures to calculate the 'warranted multiple' based on comparable companies. See below for CMG

 

 

Exhibit 2. Best Fit Valuation Regression: Restaurants: In this example, we look at Chipotle (CMG) relative to its peers in the restaurant category. While the stock has underperformed significantly, when compared against profitability, growth and capital structure factors, it does appear undervalued if they can meet current expectations. 

Share on Twitter
Please reload

Recent Posts

July 10, 2019

May 23, 2018

Please reload

Archive
Please reload

Follow Us
  • Twitter Basic Square
sales@equitydatascience.com
205 East 42nd St.,
New York, NY 10017
(646)-838-4107
EDS and EQUITY DATA SCIENCE are trademarks owned by Equity Data Science, Inc.  All rights reserved.  © 2016-2017.