The most important aspect of interpreting news and its impact on equity markets is the determination of the market's expectations for that news. In the financial world, this is commonly referred to as the "market discount mechanism". The correlation between equity markets and news is pretty clear. Expected news has little impact on stock prices while unexpected news, especially when pertaining to news that are considered likely to significantly impact the future earnings of a given company, may have an immense impact.
As a consequence, short-term traders can benefit from closely monitoring real-time financial newswire services from Dow Jones, Reuters, and Bloomberg as they are excellent gauges of current sentiment towards potential news events. Being aware of events and expectations allows traders to be fully prepared for, and profit from, the discounting of potential market moving events. Applying news analytics as part of an event driven trading strategy makes it possible to "listen in" on the news in a fully or semi automated fashion.
Event-driven trading is a fundamental based methodology that attempts to exploit the volatility associated with economic releases, political announcements, and corporate events. Often, it can be difficult to determine the effect of news on stock price movements only considering the type of event. Therefore, to better understand the likely direction of the stock price, it is useful to consider the news sentiment built up towards the scheduled corporate announcement; thereby making it possible to adhere to the old Wall Street adage of "trade on the rumor, and sell on the news".
Incorporating news analytics into one's trading model environment makes it possible not only to form structured opinions about scheduled events including earnings announcements, but also to trade on unscheduled events such as executive appointments, product recalls, lawsuits, or layoffs; which often have significant market impact. Such news events can be detected in real-time with millisecond latency, with trading decisions made based on the story sentiment as it relates to the entities being mentioned. In previous blog postings, I shared some research on the stock price impact of unscheduled news events on not only hours ahead, but as much as 60 days following the event.
Wednesday, September 30, 2009
Thursday, September 17, 2009
Construction of Simple Investment Strategies Using News-Based Market Sentiment Indices
Recently, I conducted a study on how to construct market sentiment indices based on news sentiment. The applied methodology was presented in a previous posting: "Construction of Market Sentiment Indices Using News Sentiment" from August 5th, 2009.
Generally, no single definition of investor sentiment exists. Still, most discussions involve (a) investors optimism or pessimism about stocks, (b) beliefs not justified by fundamentals, or (c) misevaluation by some investors. Since investor sentiment is not directly observable, researchers have employed various measures of sentiment. Proxies include the closed-end fund discount, mutual fund redemptions, the volume of initial public offerings, ratio of odd-lot sales to purchases, consumer and investor survey data, technical indicators, and the proportion of fund assets held in cash. In addition to these sentiment proxies, a range of economic indicators are available of which some are considered part of different governments official list of leading indicators.
The indicators that make reference to sentiment, are mainly generated from surveys and address consumer sentiment. The release schedule for these indicators are often either monthly or quarterly, making them less timely than the news sentiment indices that I suggest as part of my research. News sentiment indices are more timely because they are continuously calculated in a consistent manner from real-time media coverage.
In the U.S., examples of sentiment-based economic indicators include the Conference Board Consumer Confidence Index, the University of Michigan Consumer Sentiment Index, and the Washington-ABC News Consumer Comfort Index. While in Europe, examples include the German IFO Business Climate Index, and the French Business Sentiment Index. All of the mentioned proxies and indicators try to capture market-wide sentiment rather than sentiment at the individual company level. I propose using news sentiment in order to capture not only market-wide sentiment focusing on specific indices, but also to use news sentiment to construct company level sentiment indices (which is the focus of my current research).
The key findings of my studies on market sentiment indices can be summarized as:
It should be noted that the lead time indicates after how long an investment decision is implemented in the market. That is, with a lead time of 95 business days, the investment decision made on the sentiment index value today is not carried until 95 business days from now. The news sentiment aggregation window refers to how far back we consider the sentiment of news stories when constructing the current sentiment index value. Both figures consider a two week investment horizon.
The above results are based on sentiment data including April 2009. To present the most recent development of the market sentiment index on the DJIA. I have included the below figures that considers data including August 2009.
Interestingly, it can be observed that the sentiment levels have improved significantly especially in the second quarter of 2009; and is currently at levels that have not been seen since mid-2006, the period before the last bull market.
While more research can be done to refine the methodology and to translate the sentiment index values into actual trading or investment signals, the above results highlights some potentially interesting and profitable relationships between news sentiment and market returns. The full results of my findings are documented in the paper Construction of Market Sentiment Indices Using News Sentiment (August 28, 2009).
Generally, no single definition of investor sentiment exists. Still, most discussions involve (a) investors optimism or pessimism about stocks, (b) beliefs not justified by fundamentals, or (c) misevaluation by some investors. Since investor sentiment is not directly observable, researchers have employed various measures of sentiment. Proxies include the closed-end fund discount, mutual fund redemptions, the volume of initial public offerings, ratio of odd-lot sales to purchases, consumer and investor survey data, technical indicators, and the proportion of fund assets held in cash. In addition to these sentiment proxies, a range of economic indicators are available of which some are considered part of different governments official list of leading indicators.
The indicators that make reference to sentiment, are mainly generated from surveys and address consumer sentiment. The release schedule for these indicators are often either monthly or quarterly, making them less timely than the news sentiment indices that I suggest as part of my research. News sentiment indices are more timely because they are continuously calculated in a consistent manner from real-time media coverage.
In the U.S., examples of sentiment-based economic indicators include the Conference Board Consumer Confidence Index, the University of Michigan Consumer Sentiment Index, and the Washington-ABC News Consumer Comfort Index. While in Europe, examples include the German IFO Business Climate Index, and the French Business Sentiment Index. All of the mentioned proxies and indicators try to capture market-wide sentiment rather than sentiment at the individual company level. I propose using news sentiment in order to capture not only market-wide sentiment focusing on specific indices, but also to use news sentiment to construct company level sentiment indices (which is the focus of my current research).
The key findings of my studies on market sentiment indices can be summarized as:
- Relevance matters. The role of the company in a given story has to be taken into account as this improves the correlation between the sentiment index and the out-of-sample stock returns by a factor 2.5+. This indicates that considering only key word matching or simply counting stories that mention companies is flawed.
- More stable investment strategies. Constructing simple sentiment-based trading rules, going long (short) sentiment index values greater (lower) than 50, seem to create more stable strategies outperforming momentum-based investment strategies based on the market index itself.
- Attractive correlation levels. For the market sentiment index covering the Dow Jones Industrial Average, correlations between the sentiment index values and the two week forward-looking returns range between 20 - 27% applying a three month news aggregation window. For the Eurostoxx50 sentiment index, the corresponding range is 9 - 15%.
- Attractive Hit ratios. Applying a three month news aggregation window, I arrive at Hit ratios ranging between 59 - 65% and 55 - 67% for the DJIA and Eurostoxx50 sentiment indices, respectively; this with Profit/Loss ratios ranging between 1.01 - 1.27, and 0.86 - 1.18.
It should be noted that the lead time indicates after how long an investment decision is implemented in the market. That is, with a lead time of 95 business days, the investment decision made on the sentiment index value today is not carried until 95 business days from now. The news sentiment aggregation window refers to how far back we consider the sentiment of news stories when constructing the current sentiment index value. Both figures consider a two week investment horizon.
The above results are based on sentiment data including April 2009. To present the most recent development of the market sentiment index on the DJIA. I have included the below figures that considers data including August 2009.
Interestingly, it can be observed that the sentiment levels have improved significantly especially in the second quarter of 2009; and is currently at levels that have not been seen since mid-2006, the period before the last bull market.
While more research can be done to refine the methodology and to translate the sentiment index values into actual trading or investment signals, the above results highlights some potentially interesting and profitable relationships between news sentiment and market returns. The full results of my findings are documented in the paper Construction of Market Sentiment Indices Using News Sentiment (August 28, 2009).
Wednesday, September 2, 2009
Identifying Events Using a Sentiment Approach
Event-driven strategies are often cited as a useful alternative alpha source for quantitative investors because returns to these strategies tend to be uncorrelated with traditional multifactor quant models. However, there have always been two major challenges with event-driven investing including the infrequent nature of such events and the lack of data to detect and capture events in a timely fashion.
In some of my previous research, see the posting Impact of News Sentiment on (Intra-day) Abnormal Stock Return from June 10 2009, I proposed constructing intra-day triggering events based on unique combinations of sentiment classifications after applying a relevance filtering. A similar study was conducted by Macquarie US Equity Research with a focus on longer-term investing, see the posting How Does The Market React To News? from June 17, 2009.
Even though it is possible to construct attractive trading and investment strategies based purely on sentiment, it may be desirable to further enhance the strategy by considering other aspects of a news story. For example, news stories can be further categorized into frequent events like earnings or earnings guidance announcements, executive appointments, analysts ratings, and less frequent (and unexpected) events like product recalls, lawsuits, layoffs, or bankruptcies. News filtering can be based on the type of event of a given story, since the expected market reaction to certain events could be very different. Using this information, it may also be desirable to decide on the traded or invested amount, thereby being more bullish on a story about a new business contract than on a story about a charity donation.
Figure 2 shows the frequency of the top 20 events reported by Russell 3000 companies as sampled from the RavenPack News Scores database. As expected, Revenues is the most common event, with over 60,000 instances from January 2005 to August 2009. This particular event category captures when a company announces sales or revenue figures. Roughly speaking, earnings related events are most common, followed by sell-side analyst upgrades or downgrades, and then by company restructuring type events (reorganizations, layoffs, executive resignations, etc.).
Interestingly, it can be observed from Figure 3 that the frequency of events evolves through time. Some events, like earnings announcements, are periodic and occur every quarter (creating a seasonal pattern), whereas others are cyclical and rise and fall with the businesses cycle. Layoffs are an unfortunate recent example.
In a recent publication by Macquarie US Equity Research a number of triggering events were constructed using an ‘event-driven’ news sentiment approach. The study considered the constituents of the Russell 3000 and was based on the RavenPack News Scores database, which in addition to measuring the sentiment of news articles also flags over 100 scheduled and unscheduled events typically reported by the media.
Using news sentiment to determine the expected direction of an event is a big advantage as the other alternative is to rely purely on the market reaction of similar stocks to the same event in the past. Figure 5 shows the average excess returns of stocks after positive sentiment events, negative sentiment events, and all events, for the Top 20 events reported in the news.
On average, Macquarie finds that excess returns, particularly in the month after the event, are well delineated by the sentiment of the event. In other words, after an event with positive sentiment, on average returns tend to be positive, whereas after a negative event, they tend to be negative. However, this is not a universal result – there are a number of events (e.g., buybacks) where the sentiment of the event appears to have no impact on next month’s returns. Overall, the results in Figure 5 suggest that using sentiment to determine whether an event is likely to have a positive or negative impact has merit.
In future postings I plan to dig a little deeper into the report looking in more detail at the market reaction to some of these events.
In some of my previous research, see the posting Impact of News Sentiment on (Intra-day) Abnormal Stock Return from June 10 2009, I proposed constructing intra-day triggering events based on unique combinations of sentiment classifications after applying a relevance filtering. A similar study was conducted by Macquarie US Equity Research with a focus on longer-term investing, see the posting How Does The Market React To News? from June 17, 2009.
Even though it is possible to construct attractive trading and investment strategies based purely on sentiment, it may be desirable to further enhance the strategy by considering other aspects of a news story. For example, news stories can be further categorized into frequent events like earnings or earnings guidance announcements, executive appointments, analysts ratings, and less frequent (and unexpected) events like product recalls, lawsuits, layoffs, or bankruptcies. News filtering can be based on the type of event of a given story, since the expected market reaction to certain events could be very different. Using this information, it may also be desirable to decide on the traded or invested amount, thereby being more bullish on a story about a new business contract than on a story about a charity donation.
Figure 2 shows the frequency of the top 20 events reported by Russell 3000 companies as sampled from the RavenPack News Scores database. As expected, Revenues is the most common event, with over 60,000 instances from January 2005 to August 2009. This particular event category captures when a company announces sales or revenue figures. Roughly speaking, earnings related events are most common, followed by sell-side analyst upgrades or downgrades, and then by company restructuring type events (reorganizations, layoffs, executive resignations, etc.).
Interestingly, it can be observed from Figure 3 that the frequency of events evolves through time. Some events, like earnings announcements, are periodic and occur every quarter (creating a seasonal pattern), whereas others are cyclical and rise and fall with the businesses cycle. Layoffs are an unfortunate recent example.
In a recent publication by Macquarie US Equity Research a number of triggering events were constructed using an ‘event-driven’ news sentiment approach. The study considered the constituents of the Russell 3000 and was based on the RavenPack News Scores database, which in addition to measuring the sentiment of news articles also flags over 100 scheduled and unscheduled events typically reported by the media.
Using news sentiment to determine the expected direction of an event is a big advantage as the other alternative is to rely purely on the market reaction of similar stocks to the same event in the past. Figure 5 shows the average excess returns of stocks after positive sentiment events, negative sentiment events, and all events, for the Top 20 events reported in the news.
On average, Macquarie finds that excess returns, particularly in the month after the event, are well delineated by the sentiment of the event. In other words, after an event with positive sentiment, on average returns tend to be positive, whereas after a negative event, they tend to be negative. However, this is not a universal result – there are a number of events (e.g., buybacks) where the sentiment of the event appears to have no impact on next month’s returns. Overall, the results in Figure 5 suggest that using sentiment to determine whether an event is likely to have a positive or negative impact has merit.
In future postings I plan to dig a little deeper into the report looking in more detail at the market reaction to some of these events.
Subscribe to:
Posts (Atom)
Posts
