Reduce risk and increase efficiency by systematically incorporating the effects of public information into your workflows. “Corporate body language,” or the changes to these important market-moving events, is one of many cues that publicly traded companies can send to the market, both intentionally and unintentionally. These non-verbal tells reveal a lot about a company’s financial health and are very costly to miss. In the mid-/large-cap universe, there is a noticeable improvement in information ratios across longer aggregation windows when using the combined signal. Within the small-cap universe, the performance enhancement is more modest; however, information ratios remain relatively stable. Another significant finding of the RavenPack study was that advancers benefit from the signal over longer time periods, while the signal for delayers decays more quickly as determined by separating out long-only and short-only strategies.
- RavenPack’s clients include the most successful hedge funds, banks, and asset managers in the world.
- This is not surprising considering these companies have lower liquidity and are therefore more volatile in nature.
- A new white paper, “Trading the Earnings Calendar,” (RavenPack, 2022), provides further research on this topic using Wall Street Horizon data, with five major findings that you should consider ahead of earnings season.
- Investors and researchers have suspected for decades that text could be used to predict markets, some trying and failing.
- Internal research within RavenPack suggests that the outcomes can be sensitive not only to the version of the GPT model they used but also to the strategy implementation.
Classifying words as either positive or negative, the researchers generated article-level sentiment scores—to highlight how news likely to be perceived as positive or negative would impact stock prices. Traditionally finance researchers and market practitioners have relied on accounting data and fundamentals to predict where the market is headed. But quarterly reports arrive slowly for a market moving at warp speed, which led researchers and traders to look for other sources of predictive information, including news. To find out if news reports could be used to predict stock prices, Ke, Kelly, and Xiu borrowed machine-learning techniques used by computer scientists, who are increasingly training machines to understand text.
Insights and Limitations in Stock Price Prediction Using LLMs
In the figure below, we illustrate the impact of the expanded VWAP on the open-strategy performance applying the same prompt as in the paper, revealing a distinct deterioration evident even with a shift to a 15-minute VWAP. https://g-markets.net/ The results also highlight variations in the open-price implementation between the March and June GPT 3.5 Turbo versions. Due to the black-box nature of the models, explaining the shift in performance becomes impractical.
Business Technology Overview
At the 1-day mark, annualized returns reach 8.0% for the mid-/large-caps and 19.7% for small-caps, with information ratios of 0.8 and 1.2, respectively. Investors and researchers have suspected for decades that text could be used to predict markets, some trying and failing. As other studies have demonstrated and the RavenPack study confirms, there is immense value in staying on top of corporate events like quarterly earnings reports and their ravenpack pricing changes. Anyone familiar with the Wall Street Horizon DateBreaks Factor or Late Earnings Report Index (LERI) knows that academic research supports the idea that companies that advance their earnings date tend to share good news on their earnings calls, while those that delay tend to share bad news. The RavenPack Earnings Dates dataset consists of Wall Street Horizon earnings calendar
change records for over 8,000 stocks globally since 2006.
Insights generated systematically from over 40,000 sources of news and social media
It looks at the number of outlier earnings date confirmations and whether companies are confirming earnings dates that are later than they have historically reported, or earlier. The historical baseline for this indicator is 100, meaning that anything above this average suggests companies are confirming later earnings announcements and below this average indicates companies are confirming dates that are earlier. Wall Street Horizon found that quarters that begin with a high LERI reading end up reporting lower S&P 500 EPS YoY growth than those with lower LERI readings. Combining the two strategies by using both price movement in reaction to earnings calendar change events and earnings announcement events proved to perform best.
We classify stock returns unexplained by the valuation model as „mispriced‟ and evaluate the efficacy of this signal. We find „mispriced‟ stocks deliver an IC of 3.8% or return of 5.1% pa, which is better than that for value factors. They also have low correlations to style factors like value and analyst sentiment. Moreover, we note that the signal works well across global regions, albeit better in larger markets. In this role, she is focused on publishing research on Wall Street Horizon event data covering 10,000 global equities in the marketplace.
Using over 40,000 sources, RavenPack provides real-time news analytics, including sentiment analysis and event data focused on business and financial applications. Data includes news and social media content, allowing for comprehensive analysis of financial markets. In a recent study, the RavenPack quantitative research team explored how changes in earnings
announcement dates can offer valuable insights about stock price moves surrounding earnings
events. The research paper provided more evidence that confirms findings from previous
studies that depict earnings delays can signal weak performance, while advancing the date may
be a sign of good news. When looking at price reaction in the 20-day post-earnings period, mean excess returns are positive for advancers and negative for delayers across all market caps.
Unstructured data is growing rapidly
Every day, we express ourselves in 500 million tweets and 64 billion WhatsApp messages. On Facebook, 864 million of us log in to post status updates, comment on news stories, and share videos. We enable our clients to quickly extract value and insights from large amounts of information. Explore News and Job Analytics with a knowledge graph across 12 million business-relevant entities. By keeping institutional investors and traders apprised of critical earnings date revisions, they
can take advantage of – or avoid – short-term volatility in a given security.
Again, this is true when looking at the calendar change long-short strategy that measures stock movement in reaction to the earnings date change before the quarterly report is released. While small-caps continue to outperform here (and decay more slowly), annualized returns and information ratios are maximized at the 10-day aggregation window. When using this long-short strategy across different aggregation windows and market caps (Figure 4), the 1-day aggregation window achieves the best results.
In our recent White Paper, RavenPack data scientists sought to test this belief by constructing strategies that bought and sold stocks when companies changed their earnings dates. We leverage RavenPack‟s news-flow database to identify corporate actions like Share Buybacks, M&A, Executive Employment, Clinical Trials, etc. that act as catalysts in either driving mean reversion or explain the persistence of stock „mispricing‟. We show that complementing the „mispricing‟ signal with corporate action news-flow helps to gain a better understanding of stock price behaviour and improves the performances of these trading strategies. Aakarsh is in charge of corporate strategy overseeing new revenue generation initiatives and investments at RavenPack.
Since 2003, RavenPack has pioneered investment-grade sentiment analysis in financial services. We do not believe in “one size fits all” and have developed multiple sentiment techniques where some leverage millions of rule sets while others use sophisticated machine learning algorithms. Another Wall Street Horizon proprietary metric that considers these important earnings announcement changes and therefore offers a view on corporate confidence is the Late Earnings Report Indicator (LERI).
When companies change the dates of their official earnings releases it has been speculated that it is because they want to delay the release of bad news or bring forward the release date for good news. Marina joined RavenPack in 2004 and is responsible for Ravenpack’s financial health and strategy by leading the finance, accounting and tax functions. She is a results-driven, finance leader with more than 15 years experience in the technology industry. RavenPack maintains a database of over 20 years of historical content that includes news and social media, industry and earnings call transcripts, insider transactions, and other regulatory filings. “The Covid pandemic has forced companies to reassess the way they monitor emerging risks,” said Armando Gonzalez, CEO of RavenPack.
Insights
Figure 3 below shows how mid-/large-cap companies have a greater reaction to delayed dates and experience more momentum on the negative leg, while the small-caps react more to advanced dates resulting in greater momentum on the positive leg. The signal, although strong, decays relatively quickly, with the difference between the average advance and delay reactions reaching a peak in just a few days. For every entity and event detected in a story, RavenPack provides advanced analytics including relevance scoring, novelty tracking, and impact analysis.
Find workforce insights from the fully indexed content of 200 million job postings.
View All General Business
We attempt to rationalise whether a stock is „cheap‟ or „expensive‟ for a reason, or „mispriced‟. We highlight the limitations of valuation models in that stock price can be driven by sentiment, which is difficult to capture, or due to errors in forecasting earnings or discount rates which limits the usefulness of valuation models. We show that identifying additional drivers of returns like sentiment, management quality, earnings visibility and leverage helps to discriminate between stocks that are „mispriced‟ and „cheap/expensive‟ for a reason.
