To back-test data, a dataset is divided into a training portion and a testing portion. The typical split of the data is 50/50 or 66/33. A model is built using the training data.
At Eagle Alpha, our analysts are always conscious of the tradeoff between improving the collection technique for a data source while still maintaining the integrity of the signal derived from that source.
Eagle Alpha’s Search Signals product consists of a series of company revenue indicators constructed using Google search volumes for a company’s product offering.
The science of artificial intelligence and alternative data isn’t fiction anymore.
Natural Language Processing (NLP) is an evolving field that asset managers are starting to use in their investment processes.
Eagle Alpha analysts have spent over three years finding the best way to use online search data to predict economic indicators. Eagle Alpha has devised its own proprietary methodology that leverages all relevant academic research, as well as accepted best practices in the field.
Eagle Alpha analysts have spent over three years finding the best way to use online search data to predict economic indicators. Eagle Alpha has devised its own proprietary methodology that leverages all relevant academic research, as well as accepted best practices in the field.
CRISIL, majority-owned by S&P, prepared a report (attached) analyzing the state of big data adoption in the asset management industry. CRISIL surveyed asset managers finding that in 2017 over 80% of their respondents were planning to increase investments in big data (figure 1).
In this series we want to bring some key Data and Software Engineering topics to our audience. A particular focus will be on cloud services which can be created and deployed just in time, meaning new teams can get off the ground quickly and use best practices.
Winton Capital Management organized a data science workshop at its San Francisco office that opened last year.
In our recent conversations with industry partners, there is one issue that comes up repeatedly – the issue of explanation, i.e. explanation of machine learning models, the associated transparency and trust.
The attached white paper was prepared by Webhose.io, a company providing access to structured web data. This paper gives an overview of technological issues associated with extracting web data from various and fragmented sources.
