Advancing investment decision-making to the next level.
This is a service that distributes investment decision support information enhanced with original added value by utilizing alternative data and advanced AI technologies.
Why Choose Us
Analytical Technology
Established know-how for analyzing and modelling data, through which valuation of firms can be made.
Data Sources
A wide variety of structured data ranging from traditional ones such as financial data in all the listed firms in Japan to alternative data such as text-based ones processed with news provided by our affiliate companies, Nikkei and QUICK.
Diverse factors and signals
Five types of factors and signals are provided, including News Sentiment Factors and Access Ranking Factors. Additional factors and signals will be added as needed.
Main Functions
- Overview of factors and signals
- Use cases
News factors are developed based on natural language processing models. The model is composed of a pre-trained model and a fine-tuned model. The pre-trained model is trained using millions of Japanese articles published by Nikkei Inc. In the fine-tuning phase, part of the parameters of the pre-trained model are re-estimated using two types of labeled data. One is a sentiment factor based on human-labeled news data (positive, negative, neutral), and the other is a quantitative factor based on daily returns.
Access Ranking Factors are factors developed based on access frequency to specific stocks on the Qr1 platform provided by QUICK. Access rankings can be utilized as a proxy variable representing investor interest.
Corporate earnings forecasts are projections of near-term profits or losses officially announced by the companies. The advantages of this data include broad coverage of nearly all listed Japanese companies, long-term historical data since 1997, timely updates, and no sell-side bias.
Consensus data is sell-side analyst consensus data compiled by QUICK and is widely known in Japan as a reliable data source. The advantages of this data include the widest coverage among Japanese equity consensus data (over 1,600 stocks), timely updates, and long-term forecasts of up to five years. Analyst consensus data is complementary to corporate earnings forecast data.
We hold approximately 100 factors, which are broadly classified into three categories: value, quality, and momentum. Creating composite factors by combining these factors is useful for building efficient investment portfolios.
User Voice
The service is utilized by overseas hedge funds.
About Us
