At a Glance: Orbit Financial Technology Ltd. – China A-Shares Transcripts
This dataset offers a single source of timely disclosures from 3,500+ China A-Shares companies in a standardized semi-structured format, without requiring resource-intensive collection and translation processes.
Despite China being one of the world’s largest economies, investing in Chinese companies presents challenges to foreign investors. Less mature governance practices and a language barrier often make it difficult to access information necessary to analyze those companies. Orbit’s China A-Shares Transcripts data feed brings transparency to company communications through semi-structured transcripts of earnings calls, broker on-site research disclosures, and online Q&As with company executives. All transcripts are provided in Chinese and English.
This dataset offers a single source of timely company disclosures in a standardized semi-structured format, without requiring resource-intensive collection and translation processes. Orbit currently covers over 3,500 China A-Shares companies, which represent 95% of all China A-list companies. History is available to 2018; Orbit plans to add an additional five years of history in the coming months.
Asset Class: Public Companies
Data Frequency: Event-driven
Delivery Frequency: Daily
History: Data available back to 2018
The Orbit China A-Shares dataset contains three types of company communications: annual and quarterly earnings call meetings, broker on-site research disclosures, and executives’ official responses on online platforms. Broker on-site research disclosures are created because brokers/investment banks are obligated to disclose findings of ad-hoc research they perform on site at companies. Executives’ official responses refer to a few online forums that allow investors to ask questions to executives of listed companies directly, to which they can respond selectively. These exchanges happen in real time and provide timely insights on listed companies.
Orbit leverages proprietary web-crawlers to collect the raw data from public websites, then runs internal processes to filter, cleanse, and standardize the communications. This process includes mapping each transcript to a ticker, extracting the relevant metadata, and translating each to English. Orbit layers automated and manual checks at every stage of the process to ensure every transcript meets their high quality assurance standards.
Through this process, Orbit delivers individual JSON files for each transcript and company disclosure in English and Chinese, which are accompanied by the key pieces of metadata, including stock ticker, event time, and event type.
Quantitative Research and Risk Management
Orbit's semi-structured transcripts in Chinese and English enable users to programmatically evaluate topics and sentiment present in company communications, including earnings calls and public disclosures that are specific to Chinese companies, such as online forums and onsite broker research.
The standardized structure of the JSON files enables users to leverage customized NLP to generate topic-specific sentiment analysis, which can be used for factor generation and is additive to quantitative strategies.
By combining customized NLP with Orbit’s transcripts and company disclosures, users can systematically monitor a universe of companies to identify notable events or potential headwinds.
Orbit’s transcripts and company disclosures bring a level of transparency to China A-Shares that previously would have required immense resources to collect, process, and standardize, on top of translating the raw sources to English. This information is now accessible to investors, enabling them to capture qualitative information present in these disclosures and enhance existing fundamental analysis, essentially eliminating one barrier to increasing portfolio exposure to the Chinese market.
The details provided above are as of September 2020.
If you have any questions or would like to learn more about any of the content mentioned above, please contact us at OFSupport@factset.com.