Jay's Asset Allocation Blog

Blog about my off-hours work on the problem of Asset Allocation including but not limited to Portfolio Optimization algorithms, algorithms and approaches for improved estimation of Asset Allocation inputs and other potentially related items.

Monday, June 30, 2008

ARPM Bootcamp and Black-Litterman Views

I just spent 3 days at the ARPM Bootcamp at the Courant Institute of NYU taught by Attilio Meucci. It was an intense 3 days of financial mathematics with the instructor taking no prisoners. The course was based on the instructor's book, and having read the book I was pleasantly surprised that I could follow along with the lectures. Still managed to learn a few things, and for the rest become a little more confident in my understanding.

I'm working on some more Black-Litterman ideas for the site, right now there is a paper TEV Sensitivity to Views in Black-Litterman Model by Braga and Natale that looks like another interesting way to measure how to measure the impact of the views. Meucci has an interesting paper, Assessing Views which describes using the Mahalanobis distance to measure the impact of the views on the returns.

Monday, June 23, 2008

Interesting Thesis on Black-Litterman

I periodically search the web looking for new and interesting papers focused on the Black-Litterman model, or asset allocation/portfolio optimization in general. I recently found
The Black-Litterman Model Hype or Improvement?
by Anisa Salomons. It is an interesting work, in that it explores how to calibrate certain parameters such as τ and δ. She works a specific model using Fama and French type factor portfolios, and then uses factor based views generated by yet another model looking at some economic statistics. She reports results for different values of the parameters and draws. There is still work to be done, but this thesis is a nice contribution to the Black-Litterman model literature.

Wednesday, June 11, 2008

Global Equilibrium

I've been working on recreating the Global Equilibrium from Chapter 6 in Litterman's book. It's been an adventure as I've been learning how to use MATLAB/Scilab at the same time. I'm finally starting to get comfortable and able to do things more quickly in Scilab than in java which is good. I will be writing up what I've done, and posting the write up and the SciLab code to reproduce results up on the website hopefully in the next few weeks. Then it will be on to building some type of interesting Black-Litterman example.