Exploring Credit Default Swap (CDS) Market Data Using Modern Data Science Techniques
[et_pb_section admin_label=”section”][et_pb_row admin_label=”row”][et_pb_column type=”4_4″][et_pb_text admin_label=”Text” background_layout=”light” text_orientation=”left” text_font_size=”18″ use_border_color=”off” border_color=”#ffffff” border_style=”solid”]
October 2nd, 2018
Title VII of Dodd-Frank Wall Street Reform and Consumer Protection Act addresses the gap in U.S. financial regulation of OTC swaps by providing a comprehensive framework for the regulation of the OTC swaps markets. The objective of this blog is to describe how to rapidly and securely analyze credit default swap (“CDS”) transaction data using cloud computing and advanced machine learning (“ML”) techniques. We obtained the CDS data from the Depository Trust and Clearing Corporation (“DTCC”).
A fundamental technology enabler for our customers is the B23 Data Platform which is a Cloud-based artificial intelligence (“AI”) engine to discover, transform, and synthesize data from a variety of sources to provide unique and predictive insights. The B23 Data Platform is used by data-centric enterprises in many different industries including technology companies, government agencies, and financial institutions to securely use the Amazon Cloud to gain insight from very large data sets.
Scope and Accomplishments of our efforts include:
Created secure Machine Learning (“ML”) Analysis Cluster in representative Customer Private Cloud
Ingested 5 years of CDS data in into the ML analytics cluster in 1 minute
Identified anomalous CDS trading activities for complex market transactions
Created established CDS compliance reporting metrics
Identified CDS market characteristics at individual products and individual series levels
Identify Anomalous Transaction Activity or Faulty Reporting in Markets
B23 investigated several transactions in the DTCC CDS data that looked peculiar. Figure-2 shows a relatively small number of transactions composed of multiple line items in the data set that exhibited anomalous market activities.
In the example above, “yellow” nodes are correction activities, and “red” nodes are cancellation activities.
Figure 2: Simple Trade with Conflicting Transactions
Walking through each set of transactions, the following activities are occurring:
A single contract is created for a value of $52M (blue node labeled “new”)
The $52M original CDS is corrected to a value of $83M, and the previous $52M is canceled
The $82M corrected CDS is corrected twice $31M, and the previous $83M is canceled twice
The $31M corrected CDS is corrected back to $52M, and the previous $31M is canceled
Finally, the $52M corrected CDS is canceled twice
In this example above, activities such as cancellations and corrections against a CDS product are occurring more than once indicating anomalous and irregular behavior. B23 was successful in analyzing more complex transactions exhibiting anomalous behavior such as the one in figure-3.
Figure 3: Complex Trade with Conflicting Transactions
CDS Reporting Capabilities
Using modern data processing systems like Hadoop and Spark in the cloud, immediate cost efficiencies can be gained in generating existing compliance reports. The following examples show how B23 data scientists were able to query CDS data (queries that only taking a few seconds) to provide basic reporting features similar to those reports on the CFTC website. Figure-4 is a report showing weekly CDS transaction volume for the data we obtained from DTCC.
Figure 4: Replicating Existing Reporting Capabilities
This transaction volume data allowed us to investigate specific versions of products such as iTraxx, as indicated in figure-5. This insight allowed us to delve deeper according to volume for specific products.
Figure 5: iTraxx Volume by Product (USD)
After observing figure-5, we wanted to know the specific market pricing, in terms of basis points, for the individual series of the iTraxx Europe product, which clearly had the most volume. Figure-6 shows how the market responds to pricing as new series are introduced every 6 months. In this case, viewing 4 years of implied prices over 8 different series.
Figure 6: iTraxx Europe market Price By Series (USD)
A similar set of reports was created for CDX market transactions, with a deep-dive on the North American Investment Grade series volume over a five (5) year period.
Figure 7: Weekly CDX Transaction Volume (USD)
Figure 8: Trade Volume North American Investment Grade by Series
B23 Data Platform
The B23 Data Platform was created as a result of the B23 founders’ unique experience working at Amazon Web Services (“AWS”) and observing first-hand the challenges organizations face when processing large amounts of data securely in the Cloud. The B23 Data Platform makes it easy for business analysts to access large datasets in minutes and with just a few mouse clicks, start analyzing that data using Cloud-scale computing and storage resources.
Our platform has been vetted from a security perspective by several of the most security-sensitive organizations in the world. It is in use by customers in the technology sector, financial services, and government. The B23 Last Mile component is a component of the B23 Data Platform and allows business analysts to automatically connect to disparate data sources in a single command, making it easy for non-technical staff to analyze complex and large data sets using familiar tools like R and Python.
In this case, we ingested five (5) years of CDS data into our data platform in one (1) minute.
To learn more please contact B23 at firstname.lastname@example.org