Valuation Adjustment API

This module provides functions for calculating various valuation adjustments in financial risk management.

Functions

torchquantlib.core.risk.valuation_adjustment.valuation_adjustment.calculate_cva(exposure: Tensor, default_prob: Tensor, recovery_rate: Tensor) → Tensor[source]

Calculate the Credit Valuation Adjustment (CVA).

CVA represents the market value of counterparty credit risk.

Parameters:

  • exposure (Tensor) – Expected positive exposure to the counterparty.

  • default_prob (Tensor) – Probability of default of the counterparty.

  • recovery_rate (Tensor) – Expected recovery rate in case of default (between 0 and 1).

Returns:

The calculated Credit Valuation Adjustment.

Return type:

Tensor

torchquantlib.core.risk.valuation_adjustment.valuation_adjustment.calculate_dva(exposure: Tensor, default_prob: Tensor, recovery_rate: Tensor) → Tensor[source]

Calculate the Debit Valuation Adjustment (DVA).

DVA is similar to CVA but represents the credit risk of the entity itself.

Parameters:

  • exposure (Tensor) – Expected negative exposure (i.e., liability) to the counterparty.

  • default_prob (Tensor) – Probability of default of the entity itself.

  • recovery_rate (Tensor) – Expected recovery rate in case of the entity’s default.

Returns:

The calculated Debit Valuation Adjustment.

Return type:

Tensor

torchquantlib.core.risk.valuation_adjustment.valuation_adjustment.calculate_fva(exposure: Tensor, funding_spread: Tensor, maturity: Tensor) → Tensor[source]

Calculate the Funding Valuation Adjustment (FVA).

FVA represents the cost of funding for uncollateralized derivatives.

Parameters:

  • exposure (Tensor) – Expected exposure over the life of the derivative.

  • funding_spread (Tensor) – The entity’s funding spread above the risk-free rate.

  • maturity (Tensor) – Time to maturity of the derivative.

Returns:

The calculated Funding Valuation Adjustment.

Return type:

Tensor

torchquantlib.core.risk.valuation_adjustment.valuation_adjustment.calculate_mva(exposure: Tensor, funding_cost: Tensor, maturity: Tensor) → Tensor[source]

Calculate the Margin Valuation Adjustment (MVA).

MVA represents the cost of posting initial margin for cleared or non-cleared derivatives.

Parameters:

  • exposure (Tensor) – Expected initial margin requirement.

  • funding_cost (Tensor) – The cost of funding the initial margin.

  • maturity (Tensor) – Time to maturity of the derivative.

Returns:

The calculated Margin Valuation Adjustment.

Return type:

Tensor

Detailed Description

The valuation adjustment module offers a set of functions to compute different types of valuation adjustments commonly used in financial risk management and derivatives pricing.

Credit Valuation Adjustment (CVA)

The calculate_cva function computes the Credit Valuation Adjustment, which represents the market value of counterparty credit risk.

\[CVA = Exposure \times (1 - RecoveryRate) \times DefaultProbability\]

Debit Valuation Adjustment (DVA)

The calculate_dva function calculates the Debit Valuation Adjustment, which is similar to CVA but represents the credit risk of the entity itself.

\[DVA = Exposure \times (1 - RecoveryRate) \times DefaultProbability\]

Funding Valuation Adjustment (FVA)

The calculate_fva function determines the Funding Valuation Adjustment, which represents the cost of funding for uncollateralized derivatives.

\[FVA = Exposure \times FundingSpread \times Maturity\]

Margin Valuation Adjustment (MVA)

The calculate_mva function computes the Margin Valuation Adjustment, which represents the cost of posting initial margin for cleared or non-cleared derivatives.

\[MVA = Exposure \times FundingCost \times Maturity\]

Usage Example

Here’s a basic example of how to use the valuation adjustment functions:

import torch
from torchquantlib.core.risk.valuation_adjustment.valuation_adjustment import calculate_cva, calculate_dva, calculate_fva, calculate_mva

# Set up parameters
exposure = torch.tensor(1000000.0)
default_prob = torch.tensor(0.05)
recovery_rate = torch.tensor(0.4)
funding_spread = torch.tensor(0.02)
funding_cost = torch.tensor(0.03)
maturity = torch.tensor(5.0)

# Calculate adjustments
cva = calculate_cva(exposure, default_prob, recovery_rate)
dva = calculate_dva(exposure, default_prob, recovery_rate)
fva = calculate_fva(exposure, funding_spread, maturity)
mva = calculate_mva(exposure, funding_cost, maturity)

print(f"CVA: {cva.item():.2f}")
print(f"DVA: {dva.item():.2f}")
print(f"FVA: {fva.item():.2f}")
print(f"MVA: {mva.item():.2f}")

Note

All functions in this module use PyTorch tensors for input and output, allowing for efficient computation and automatic differentiation when needed.