<< Back to articles list With a Challenge, the user has the ability to trigger a speedy OTCStreaming Data Update. The user is expected to be part of the data quality challenges: Accuracy, Consistency and Transparency.

If you’ve ever tried to contact a data service help desk, you know it is very hard, even impossible, to correct one data point from among tens of thousands supplied by a data provider. The data provider has its own data cleaning process and users are not part of it. This can be a real source of frustration.

OTCStreaming would like to offer “zero error” time series. Although we use reliable sources of data like SDRs which are used for over-the-counter trade information dissemination, errors are still frequent. For instance the standard coupon of a contract can be reported instead of its traded spread. Any time-series analysis will be affected by such outliers. Any quantitative analysis will give spurious results. The data scientist can design his bespoke filter to remove suspect data but OTCStreaming introduces the concept of “Challenges”. With a Challenge, the user is able to share a concern on a data point and can trigger a data inquiry. With a Challenge, the user is now involved in the data quality process. With the help of its community, OTCStreaming strives to provide “zero error” time series.

The OTCStreaming’s Data Moderation Team (DMT) is in charge of reviewing all Challenges. The reader should refer to our Data Manifesto to get further insight on our data quality process.

To keep the data quality process transparent, the DMT’s own Challenges will follow the same stream when it has spotted an error. Exhaustive lists of DMT’s Challenges and users’ Challenges are available online.

Again, with transparency in mind, no data is modified. The system appends an amended data entry with a specific tag. Users can always go back to the raw source of the data if they want to, especially to test the quality of DMT’s work!

<< Back to articles list