Re: [Quantlib-users] Asset Swap Calculation Error with IBM 1.25 1/29/27 (Settlement Days Issue)

[Aaron De la R. <aar...@gm...>]

OK. Let me check it out. I'll be back later.
________________________________
From: Kenji Ogawa <for...@gm...>
Sent: Monday, October 27, 2025 9:37 AM
To: Aaron De la Rosa <aar...@gm...>
Cc: Qua...@li... <Qua...@li...>
Subject: Re: [Quantlib-users] Asset Swap Calculation Error with IBM 1.25 1/29/27 (Settlement Days Issue)

Hello,
Please find attached the Jupyter Notebook code.

When the parameters are set as " settleCRV, settleBND = 1, 2 ", the calculations run correctly. However, changing them to " settleCRV, settleBND = 2, 2 " produces abnormal results.

It appears that the curve date needs to be set one day earlier than the bond settlement date.

regards,
Kenji



2025年10月28日(火) 0:14 Aaron De la Rosa <aar...@gm...<mailto:aar...@gm...>>:
Could you share the entire code. So, I can fix it.

Regard

Aaron
________________________________
From: Kenji Ogawa <for...@gm...<mailto:for...@gm...>>
Sent: Sunday, October 26, 2025 6:34 PM
To: Qua...@li...<mailto:Qua...@li...> <Qua...@li...<mailto:Qua...@li...>>
Subject: [Quantlib-users] Asset Swap Calculation Error with IBM 1.25 1/29/27 (Settlement Days Issue)


Below is the code that calculates the asset swap for IBM 1.25 1/29/27, traded on April 25, 2024, at a price of 93.95.

When using settleDS = 3, the results are correct:
legNPV1 = 95.179 and fairSPD = 0.359%.

However, when using settleDS = 2, the calculation produces abnormal results:
legNPV1 = 100.926 and fairSPD = -1.81%.

Since the bond should settle on T+2, this discrepancy is problematic.
Could someone please help fix this error?