Risk-Neutral Bond Pricing When Agents Are Risk Neutral
Article Quant Q&A · Author: qp212223
Summary
The document raises a question about a bond-pricing argument in Duffie and Lando’s model. The model is described as specifying an asset that follows geometric Brownian motion and pricing a bond using a restricted information filtration, without an explicit change to a risk-neutral measure. The question asks why this is valid.
Key ideas
- The cited explanation relies on the model’s assumption that agents are risk neutral.
- Under that assumption, the answer identifies the physical and risk-neutral probabilities as equal.
- The excerpt does not develop the bond-pricing derivation or discuss other model assumptions.
Tags
Full text
# Question about (lLack of) Risk Neutral Bond Pricing in Duffie & Lando (2001)
# Question about (lLack of) Risk Neutral Bond Pricing in Duffie & Lando (2001)
I have been reading this famous paper of Duffie & Lando (2001) and I have a question regarding how they calculate the price of a bond (the reader of this post will not have to dive deep into the details, but it would certainly be helpful to be familiar with the paper). In particular, they fix a probability space on which an asset $V_t$ follows geometric Brownian motion with drift $\mu \equiv m + \sigma^2/2$ and volatility $\sigma$ and compute the price of their bond (under a filtration which is a strict subset of the filtration of $V$) without changing to the risk neutral measure (see equation (29) and equation (26), specifically). Why is this valid, or am I just missing something entirely?
## Answer by Achrbot (score 1)
https://quant.stackexchange.com/a/78461
In the model setup, they state that all agents are risk-neutral, which would mean $\mathbb{P}=\mathbb{Q}$.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.