Remove special initial guess for spread discount curves#2495
Merged
lballabio merged 1 commit intolballabio:masterfrom Mar 23, 2026
Merged
Remove special initial guess for spread discount curves#2495lballabio merged 1 commit intolballabio:masterfrom
lballabio merged 1 commit intolballabio:masterfrom
Conversation
The current logic is to use 0 spread as the initial guess. Revert this to the logic used for outright curves, which guesses 5% spread. It turns out that 0 spread may be very close to the final result, which creates issues for the global bootstrap. When the initial guess is too close to the final answer, gradients become too small and the LevenbergMarquardt algorithm used by the global bootstrap does not have a good direction to move into. Counterintuitively, this makes convergence slower than when starting from a much further initial guess like 5%. In our production setting when starting from 0 spread, we saw curves take multiple seconds to build or fail with MaxIterations exceeded, but converge in 100ms when starting from 5% spread.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The current logic is to use 0 spread as the initial guess. Revert this to the logic used for outright curves, which guesses 5% spread. It turns out that 0 spread may be very close to the final result, which creates issues for the global bootstrap. When the initial guess is too close to the final answer, gradients become too small and the LevenbergMarquardt algorithm used by the global bootstrap does not have a good direction to move into. Counterintuitively, this makes convergence slower than when starting from a much further initial guess like 5%. In our production setting when starting from 0 spread, we saw curves take multiple seconds to build or fail with MaxIterations exceeded, but converge in 100ms when starting from 5% spread.