OSTTRA has expanded its triResolve portfolio reconciliation service with the launch of the FX Snap Time Prediction Model.
The FX Snap Time Prediction Model is designed to identify and flag valuation discrepancies that arise when counterparties capture market rates at different points in time.
The new prediction model uses historical trade valuation data to form a time-series analysis.
By combining this data with the prevailing market rate, it identifies when a trade’s mark-to-market (MTM) value was determined, then applies the same analysis to the counterparty’s side of the trade to calculate whether the difference is down to timing rather than pricing.
Carl Thornberg, head of optimisation and analytics technology at OSTTRA, said: “The FX Snap Time Prediction Model helps our customers solve a long-standing market challenge.
“By isolating valuation discrepancies caused by snapshot timing, we are helping market participants eliminate false positives, allowing them to focus resources on real risk.”
According to the Uncleared Margin for OTC Derivatives survey, these mismatches are among the most common causes of margin disputes, as MTM values frequently diverge depending on when each side takes its snapshot.
To address this, the new system collects and analyses data 24 hours a day, five days a week, to cover global FX trading.
Read more – BNP Paribas takes strategic stake in OSTTRA
Warren Rees, executive director, CPG and collateral operations, digital and platform services at JP Morgan, said: “FX timing mismatches have historically been a significant operational hurdle.
“By distinguishing discrepancies attributable to snapshot timing, operations teams can more efficiently triage exceptions and escalate where further review and oversight may be required.”
Following a four-week testing phase with nine major firms, the FX Snap Time Prediction Model is now live for all clients.