Estimating Collateral Settlement Time with Granger Causality

Summary

The document discusses estimating the settlement period for collateral in non-cleared OTC derivatives, within counterparty credit risk modeling. It places settlement time inside the margin period of risk: while collateral-related cash flows accumulate, a default may occur; once recognized, liquidation begins. Variation margin and initial margin are presented as tools intended to mitigate exposure over that horizon.

The proposed estimation uses Granger causality between theoretical collateral and the actual collateral available for each counterparty. The questioner suggests taking the average time lag found across counterparties as an estimate of settlement time. This is a proposed method in a request for relevant research, not a demonstrated result. No data, test specification, validation, or evidence that the estimated lag corresponds to operational settlement time is supplied. The document also leaves open how to interpret causality-test lags and distinguish settlement delays from other sources of timing differences.

Key ideas

• Settlement time is described as part of the margin period of risk for collateralized OTC derivatives.
• The period concerns cash-flow accumulation before default recognition and liquidation.
• The proposed method tests whether actual available collateral is Granger-caused by theoretical collateral.
• The average lag across counterparties is suggested as a settlement-period estimate.
• No empirical validation or test details are provided to establish that the lag measures settlement time.

Tags

# How to estimate quantitatively the settlement period?


# How to estimate quantitatively the settlement period?












The context of this question is Counterparty Credit Risk. In particular, the modelling of collateral for non-cleared OTC derivatives.

Regulators require collateral amounts, such as Variation Margin and Initial Margin, to be exchanged with a counterparty throughout the lifetime of a portfolio of trades.

This collateral should mitigate the risk of a counterparties' default over some horizon, e.g. 10 working days, called the margin period of risk (MPOR).

The settlement period is part of that MPOR, and corresponds to the time during which cash flows are acummulated by either side. This period ends when a default event is recognized, after which the liquidation procedure starts.

The question is, do you know of any available work on how to quantitatively determine this settlement period over counterparties?

So far we have approached this using Granger causality test. In short, we check whether the actual available collateral per counterparty is Granger-caused by the theoretical collateral. Then we use the average time-lag over counterparties, resulting from this test, as an estimation of the settlement period.

Kind Regards,

Nicolas

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.