Equities
November 14, 2017 (Updated version February 2023)
In this study, we demonstrate how RavenPack Analytics can be predictive of stock performance after neutralizing the exposure to MSCI Barra risk factors.
The Barra Risk Factor Analysis is a risk model developed by MSCI, an American finance company. It incorporates over 40 data metrics, such as earnings growth or share turnover. It is used to measure the overall risk associated with a security relative to the market, around three components:
For the sake of simplicity, when evaluating how efficient a trading signal is, researchers often use excess returns. This is equivalent to assuming just a single risk factor (the market) with constant factor exposure (𝛽 = 1) – inspired by CAPM (Capital Asset Pricing Model). In reality, there are several other risk factors and exposure differs not only across time, but also across assets. This can result in incorrect conclusions about signal performance, including when it comes to volatility and signal return.
To effectively isolate the effects of a given signal, we need to account for a broader set of risk factors. Also, we need to make sure that any observed performance cannot be explained by known risk premia - this is the margin by which the projected return on a risky asset is expected to beat the return on a risk-free asset.
Sentiment analysis, a quantification of sentiment based on the analysis of written texts and/or expert consensus on the potential impact of events, is increasingly being seen as an alternative source of information for enhancing risk models.
In this paper, we are introducing a risk model to examine the performance of RavenPack sentiment signals after taking into account the contributions from the market, industries, and styles such as value, momentum or size, among others.
We move beyond excess returns to consider the specific returns (as defined by several distinct MSCI Barra risk models) in a daily strategy which improves risk-adjusted performance driven by both higher returns and lower volatility.
This is done as a convenient shortcut to a factor-neutral portfolio implementation. In this context, we evaluate the effect of applying a regional model vs. a global model (USFAST/ USMED vs. GEM3), and faster vs. slower factors (USFAST vs. USMED).
To create our sentiment signal, we take a 1-day average of RavenPack’s Event Sentiment Score (ESS), filtering on highly relevant and novel events. The resulting sentiment indicator is then used as input to a simple linear regression model with the dependent variable being the MSCI specific returns .
Experimental Stats. Comparison between MSCI specific returns and excess returns for a set of statistics. Results are shown for USFAST model for Large/Mid-Cap and Small-Cap. Source: RavenPack, MSCI, November 2017.
Please use your business email. If you don't have one, please email us at info@ravenpack.com.
We will process your personal data with the purpose of managing your personal account on RavenPack and offering our services. You can exercise your rights of access, rectification, erasure, restriction of processing, data portability and objection by emailing us at privacy@ravenpack.com. For more information, you can check out our Privacy Policy.
Your request has been recorded and a team member will be in touch soon.
RavenPack is introducing a multidimensional approach to news analytics that captures the evolving complexity of news, enabling traders to build more dynamic and transparent strategies.
New research using RavenPack Job Analytics to quantify tech adoption in hiring posts proves that companies hiring for novel tech skills outperform peers from an investment perspective.
Our multi-asset allocation strategy based on inflation nowcasts and RavenPack sentiment analytics outperforms the S&P 500 and stabilizes volatility. With real Sharpe ratios of up to 1.25 and a risk control approach, it delivers a compelling inflation hedge for investors.