Choosing Interpolation for Discount Factor Curves and Forward Rates
Summary
The document raises the choice of interpolation method for discount factor curves. It states a preference for log-linear interpolation and suggests that the reason may be its relationship to avoiding arbitrage in forward rates. The central issue is whether that preference is supported by theory, literature, or practical test cases.
No derivation, cited evidence, comparison of interpolation methods, or worked example is provided. The claim about arbitrage is presented as the author’s understanding and is posed for discussion, so it should not be treated as a demonstrated result. The question is useful to fixed-income curve builders because interpolation affects the implied shape of discount factors and forward rates between market-observed maturities. Evaluating a method requires specifying the curve setup and the meaning of arbitrage being tested; the document itself does not establish that log-linear interpolation is universally preferred or sufficient to rule out arbitrage.
Key ideas
- The document asks how to select an interpolation method for discount factor curves.
- It identifies log-linear interpolation as a commonly favored approach and links it to forward-rate arbitrage concerns.
- The document provides no supporting references, tests, or proof for that proposed advantage.
- Interpolation choices should be evaluated in the context of the curve and arbitrage conditions being considered.
Tags
Full text
# How to determine a correct interpolation technique for discount factor curves? # How to determine a correct interpolation technique for discount factor curves? I think Log-Linear interpolation of discount factor curve is most preferred. I think this is due to the fact of not having any arbitrages in the forward rates. Can someone share some literature or some test cases where this is prominently evidenced and some supporting articles if any?
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.