Back to all articles

Missing Data in Position Limits - How we solve one of the industry’s biggest problems

3 mins
Posted on Jun 14 2021 by Liam Driscoll

How to overcome missing data for MiFID II and CFTC Position Limits for derivatives.

They say that data is the new gold. However, under the surface of the financial services industry, it tends to more accurately resemble cheap scrap metal. The quality is hugely compromised when it’s left incomplete, preventing institutions from leveraging the full potential of their data assets. At FundApps, we're trying to do our bit to help; our Missing Data functionality helps our clients identify items of missing or incorrect data and assists with a speedy resolution that ensures accurate monitoring of Position Limits.

warning sign

Where does position limit data come from?

There are a number of steps required to monitor Position Limits. Throughout the process, our service gathers data from a number of different sources, mapping it together behind the scenes to calculate the proximity of positions held by our clients to limits set by regulators and exchanges. This data is fed to the rules engine, which is our proprietary technology performing the magic to calculate results.

To start, limit data is retrieved from FIA Tech. This includes the limit itself as well as instructions for limit calculations, which we translate into machine-readable commands to be interpreted by our service. The information we grab relates to contract aggregation as well as the set of different rules provided for each contract. For example, one contract may have multiple rules stating whether the limit should be monitored on a net long/short basis, or alternatively on a gross basis. 

Spot limit instructions require trading calendar information, so data for these are gathered from the exchanges themselves. The final piece of the puzzle is the position file sent to us from our clients, which contains their relevant derivative positions. This is fed into the rule engine to determine clients’ results (i.e whether any positions breach or are close to applicable limits).

What causes missing data for position limits?

During the rule execution process, clients’ data is used to evaluate the properties required for the predefined functions that make up the rules in the rule engine. Think of this step as the point that uses raw values from client data as inputs in the instructions established by regulators and exchanges. At this stage, the evaluation will fail and cause missing data if the contract aggregation goes wrong or if the value required for the evaluation cannot be calculated. For example, if a user uploads a file without a 'Delta' property, this may cause missing data as the property is used for the calculation of some specific contracts.

Most properties will be caught during the validation phase of a position file upload. However, items that aren't required at the time of upload e.g OpenInterest that isn’t always used, or values provided that don't map correctly to contract codes from FIA Tech or exchange calendar data can cause missing data.

Monitoring missing data for position limits

As a first step, we've created a missing data table to highlight values provided by the client that cause the property evaluation to fail. This allows clients to identify data that either can't be retrieved from the portfolios section or the input file. 

Missing Data table

​Screenshot of the Missing Data table in our Position Limits service

The table above allows users to see how many rules are affected by each missing data item, and exactly which rules are affected. This is also used to calculate how many portfolios have been impacted (based on portfolios' rule assignments), including the individual assets within these portfolios.

Using all of this information, we determine and display the number of results affected by each missing property.

Why is this helpful?

All of this information may be great, but it's even better to know what to do with it. That's why we've provided solutions for each item in the table.

Depending on the property type, the user will simply be prompted to provide the value in either the portfolios section or the input file. Alternatively, the user has the option to create a data override for the property, essentially allowing them to create a dummy value.

Reducing missing properties improves the quality of data, and as a consequence helps clients improve the accuracy and integrity of their results! Want to find out more about our missing data functionality, or see what else our automated Position Limits service can do? Then get in touch!