Achieve High Capacity Investment Strategies

RavenPack | September 21, 2016

We show that supply-chain and competitive landscape information can help create more scalable portfolios from infrequent or sparse signals.

Executive Summary:

Investment Strategies

In this research, we use an earnings sentiment indicator as a signal, constructed using RavenPack Data. We find value in propagating the signal from the source companies to their suppliers and competitors.

Here are some of the highlights

  • We achieve significantly bigger and more scalable portfolios – increasing portfolio size by up to 6x.
  • Our observed performance improvements scale with both higher AUM and longer investment horizons with higher IRs from 1 to 21 days, higher annualized returns from 1 to 10 days, and lower volatilities from 2 to 21 days
  • With an AUM of $100 million, Information Ratio improves by up to 55% and up to 85% when assuming AUM of $200 million

Investment Strategies



Investment Strategies



1. Introduction

Uncovering business relationships among companies can help build profitable investment strategies. No company operates in isolation; they are all part of an ecosystem consisting of suppliers, competitors, customers, and partners, also known as economically-linked companies. Understanding a company’s business relationships informs investors about their business opportunities and risk exposure. For example, looking at the sentiment of one company can tell you how the market might perceive its suppliers or competitors, and this can be exploited in order to get larger and more scalable investment portfolios.

Portfolio size can be very important. When evaluating a strategy, its capacity is usually overlooked but its performance might be considerably affected by the invested capital. For instance, strategies involving small cap companies can appear very profitable, but their performance can deteriorate very fast for higher levels of assets under management (AUM). The main reason being that small cap companies are usually associated with more stringent liquidity constraints, which makes them more difficult to trade without moving their stock price.

In a previous study, we proved how earnings sentiment data can provide an edge over earnings consensus estimates. Nevertheless, a strategy built on top of a potentially sparse signal like earnings has some drawbacks including small portfolio sizes and limited capacity. However, since news like earnings guidance or other earnings related news can provide direct information about the future growth prospects of a company, we can also expect that it will impact those economically-linked companies that depend on it. This offers an opportunity to increase portfolio size by focusing on companies in its network.

In this paper, we investigate whether companies with earnings sentiment propagate its effect onto their suppliers, customers, or competitors, and thereby improves the ability of the signal to provide attractive performance when AUM increases. We show that the spillover strategies, combined with the original, allow to trade considerably larger portfolios. This translates into an overall strategy that is more diversified and less liquidity constrained, thus able to absorb higher levels of AUM and provide better performance in terms of Information Ratio as AUM scales.

To evaluate the scalability of our proposed signals, we develop a backtesting framework in order to assess the marginal trade-off between risk and return as more capital is allocated towards the strategy. Specifically, this is obtained by exploiting minute data on trading volume and price to impose a set of liquidity constraints, and hence adhere to the price-taker hypothesis, where the investor’s own trading has no impact on the market.

The paper is organized as follows: Section 2 describes the different data sources that we use; Section 3 follows with a brief introduction of our earnings sentiment indicator, as well as the methodology adopted for propagating the signal across the supply-chain and competitive landscape, and for evaluating strategy scalability. Results are provided in Section 4; and Section 5 presents our general conclusions.

2. Data Description

In this section, we provide an overview of the datasets that we use in our research. We describe RavenPack’s event taxonomy, which we use to derive the signal we call the...

Click here to continue reading the "Achieve High Capacity Investment Strategies By Trading Economically-Linked Companies" White Paper

Request White Paper

Request a Trial

Fill out the form below and see RavenPack in action.