<t>Andy:You are asking a very important question that is very under-appreciated: how do I match multiple moments of historical returns, at multiple holding periods, while reflecting some of the long-memory or regime aspects associated with extreme environments.The first important step in addressing ...

It is the Barclay's Capital Aggregate Index.Bloomberg ticker LBUSTRUU.

It's the replacement for Lehman Agg and is available from a variety of sources.It is an investment benchmark for many fund managers, which makes it a point of interest.

Looking at daily returns from S&P Total Return and BarCap, I see persistently negative correlation for a variety of time periods and horizon lengths.Can someone explain structurally why this should be so?

<t>OLS Regression assumes that there is a constant (but noisy) relationship that you are attempting to discover.A rolling regression pre-supposes that the relationship changes over time. PRINCIPLE: Whenever you are trying to have it both ways, you are asking for trouble.Which is it (constant or chan...

<t>Hills1234:Your originally wrote QuoteI have a high dimensional dataset...it is a 250 by 58 matrix of uniform observations. I have fitted various copula models to it including Student T, Gaussian, Archimedean, Vine copulas, etc. Now I would like to assess the fit. I have obtained the classical mea...

<t>Quote--------------------------------------------------------------------------------Originally posted by: DavidJNThey seemingly failed to grasp the implications of the no-arbitrage pricing revolution.--------------------------------------------------------------------------------So you discover ...

<t>The primary difficulty one faces in evaluating insurance models is that you have four model frameworks for four very different purposes.1. Traditional financial models for valuing and managing systematic, liquid financial market risks.2. Traditional actuarial models for valuing and managing diver...

Will do. Thank you Martinghoul!

Looking for a REFERENCE BOOK on inflation models and pricing inflation-linked securities. Have looked at Brigo-Mercurio. Anything better out there?

Just pulled down the Rebonato paper and read it.Indeed helpful and it explictly suggests that no-arbitrage is overly confining for many applications.Thank you!

I agree the real-world model has to be "realistic" and not allow outrageous forward rates.The the "internal consistency" of HJM conditions leads directly to no-arbitrage.I am asking whether looser conditions than HJM might be reasonable for some applications.

<t>I am building a rate model to evaluate risk metrics, like what is my 99%-ile worst portfolio performance over an investment horizon. It will NOT be used for any form of market pricing.This model is very realistic from multiple perspectives but is not arbitrage-free. The drift functions are not re...

<t>EXTREME DANGER: If you use classical time series approaches, the mere fact that you have a finite dataset will lead to artificially inflated indications of mean reversion.Better to (i) Use something like Hurst Exponent (and there are other more complex measures) to determine the existence of reve...

See Revuz and Yor, "Continuous Martingales and Brownian Motion"