Online and social media content can be used to analyse consumer trends and sentiment. This data can be used to examine fundamental issues that may impact consumer attitudes to a company or brand.
In this case study, Eagle Alpha investigates China datasets and explores the insights that have been found.
Using web crawl data investors can monitor regular price, markdown and promotional discounting trends across US apparel retail companies. The data provided an early read on improving fundamentals at Urban Outfitters in late 2017 and through 2018.
Using RE Analytics data on Eagle Edge investors can monitor pricing trends by channel and by geography for luxury goods companies. This data would have helped inform investors of the positive fundamentals for Burberry in the first half of 2018.
This vendor uses a combination of Artificial Intelligence and rule-based Natural Language Processing to extract unique insights from large volumes of written and spoken words. The dataset contains a grouping of 400 events into 18 high level categories.
This vendor provides data and quantitative trading strategies based on emotions in web text data. The company aggregates articles and messages (social media, social trading, blogs, forums, news) in 8 languages.
We specialize in ad hoc / one-off web scraping projects as well as data subscriptions and large-scale broad crawls.
Eagle Alpha correctly predicted GoPro results using web crawled and online search data.
This company creates datasets from a broad array of public online sources, capturing ephemeral information on the products, operating markets and labor markets of over 400,000 global companies across sectors.
This data can provide an insight into long-term trends, as well as the most recent trading performance, and is based on metrics such as average selling price (ASP) and share of “bestselling” products in a category, as ranked by the website provider.
On May 15th, 2018, Eagle Alpha organized an online meeting attended by twenty-one asset management firms to discuss the topic of testing alternative datasets for quality.
This dataset is generated by the XYME framework using a machine learning model ensemble containing more than 200,000 models (approximately 500 models for each ticker).
