I recently read about the book (Risk and asset allocation) written by Attilio Meucci and I found those statistical modeling and inference methods quite robust in my point of view, although there are always rooms to improve as the version of that book dated back to 2005.

I searched quite a long time but I could not find some other books regarding financial statistical modeling and inference as good as this one. Does any of you have some other recommendations?

  • 4
    $\begingroup$ Although I have not read the book, however I managed to briefly look at the chapters and introductions. I believe that mathematics and statistical analysis is not the right approach to asset allocation. Projecting equity and fixed income securities never works and have never worked. I see that asset allocation should be done based on macroeconomic analysis, political analysis and its impact on fixed income and equities along with broad based valuation analysis of all asset classes. $\endgroup$
    – user28909
    Jan 1 at 14:25
  • 1
    $\begingroup$ and yet hierarchical risk parity, nested clustering optimization, and even the inverse-volatility portfolio are statistical (micro) approaches to asset allocation that do work, yet rely on no macroeconomic input whatsoever. Besides various GARCH-based covariance forecasting estimators, option-implied volatility has been shown to project securities quite well during asset allocation $\endgroup$
    – develarist
    Jan 2 at 13:16
  • $\begingroup$ @develarist While it is true that risk-parity, inverse-volatility portfolios and the likes, do not incorporate any $statistical$ macroeconomic input/variables, they are still very much affected by macroeconomic drivers/events. Therefore, when constructing portfolios based on statistical asset allocation, you often rely on macroeconomic analysis to forecast changes in regimes, that will affect the diversification of the underlying financial assets in your portfolio. Thus, both you and the above answer are "correct" and often a combined setup is ideal for "superior" portfolio construction. $\endgroup$
    – Pleb
    Jan 2 at 14:34
  • $\begingroup$ affected or not affected, they do not require, and can work without, macro input $\endgroup$
    – develarist
    Jan 2 at 22:30

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.