Using Parsing to Obtain Alternative Financial Data
Demand for alternative data is growing due to its accessibility and scale. Hedge funds and investors are increasingly incorporating alternative datasets into their workflows to support investment decision-making. Among these, web data consistently proves to be the most powerful and popular.
Web data is the number one source of alternative data, helping the world’s best asset managers gain insight into market opportunities and arming them with the understanding needed to act quickly and develop positions thoroughly. When it comes to extracting high-quality data from the internet, there is virtually no limit to the type and amount of data available.
In this guide, we will cover several types of web data available for parsing and how they can be used to inform the investment decision-making process.
- Product data
- SEC filing data
- Product reviews
- Company news
- Sentiment analysis
Product Data
Collecting product data from complex online marketplaces is a challenging task, but it provides enormous insight into a range of factors critical for assessing a company’s fundamentals and stock performance.
This data opens up profitable opportunities for investors when determining market orders and positions, as well as providing insight into long-term trends.
SEC Filing Data
Diving into the lengthy and often obscure pages of company documents filed with the U.S. Securities and Exchange Commission can lead to surprising investment discoveries. It provides investors with high-quality, reliable information—data that already meets the U.S. government’s strict standards.
Now, thanks to parsing information from SEC documents, the discovery process familiar to investors can be repeated tens of thousands of times across an unlimited number of documents, uncovering valuable alpha particles in previously unseen places.
Product Reviews
To accurately predict how a company’s stock will perform, it is essential to understand how key products are tracking in the market.
Unfortunately, the lag between company metrics and quarterly earnings can reduce the usefulness of these reports for real-time analysis. Product reviews can allow investors to actively gather information about the product lifecycle and make more timely assumptions about company revenue.
Company News
Not only the company’s own announcements, but also the frequency of mentions on social media and business content providers yields extremely useful data on the company’s trajectory.
Sophisticated algorithmic traders can integrate this type of data into their processes to ensure major news events are accounted for when executing trades, and that is just a fraction of what is possible when this data type is incorporated into an investment engine.
Sentiment Data
Some time ago, Bloomberg stated that access to the Twitter feed offers one of the largest alternative datasets for investors seeking alpha. New developments in behavioral economics show that “collective sentiment derived from large Twitter feeds” can predict Dow Jones movements with a striking accuracy of 87.6%.
By extracting sentiment data from the web, investors can make timely and accurate decisions in an ever-changing market.
Modern data sources span a huge variety, from those we have already discussed to geolocation data, email receipts, and even satellite imagery. The possibilities seem endless, but sophisticated investors will achieve the best results by combining machine learning, human analytics, and large, high-quality alternative data sets.
More and better data means your investment decision-making process delivers greater value, more consistently. Moreover, adapting to the new practice of using alternative data ensures your models are aligned with the upcoming data-driven transformation of virtually every business sector.