111 reputation
5
bio website
location
age
visits member for 1 year, 8 months
seen Nov 22 '13 at 14:40

Mar
1
comment Optimization procedure for entropy pooling
Thanks John, I compared the results with the normal distribution, and the numerical solution indeed does not match the analytical solution when I apply extreme views on expectation or volatility. For those interested, the solution is in section 4 of the "Fully flexible extreme views" paper, a Gauss-Hermite grid solves the problem.
Nov
27
comment Does entropy pooling apply to distributions with time-varying drift?
In this case it's possible to simulate the distribution of log returns if I fix my investment horizon, then my drift becomes just a constant at the projected horizon. I actually have the exact case that you describe, where I have an unsustainable regime that will change at some critical point (incorporated in the model) and then reverts back to a steady state. I am curious as how you would set up a dynamic framework, I'm not sure I understand this part correctly: "treating the distribution at every horizon as one distribution", it's not the same distribution at every horizon right ?
Nov
27
comment Does entropy pooling apply to distributions with time-varying drift?
I also think it's possible, however it didn't show clearly in his paper that his model parameters (normal or fat-tailed) could be time-varying. I can send you the paper if you want to have a look at it.
Nov
27
comment Does entropy pooling apply to distributions with time-varying drift?
I had the same concern that we need a market invariant for asset allocation, but from what I see in the Meucci paper "The Prayer", invariance is needed to project the invariant distribution to the investment horizon. In my case however, I already have the projection and the pricing step at the investment horizon with my time-varying drift. Provided I fix my investment horizon, my drift becomes known. I don't think the quest for invariance should be something to always strive for, since there is large evidence that bubble regimes, for example, typically break the invariance.