Are you planning to be financially independent as early as possible so you can live life on your own terms? Discuss successful investing strategies, asset allocation models, tax strategies and other related topics in our online forum community. Our members range from young folks just starting their journey to financial independence, military retirees and even multimillionaires. No matter where you fit in you'll find that Early-Retirement.org is a great community to join. Best of all it's totally FREE!
You are currently viewing our boards as a guest so you have limited access to our community. Please take the time to register and you will gain a lot of great new features including; the ability to participate in discussions, network with our members, see fewer ads, upload photographs, create a retirement blog, send private messages and so much, much more!
Give me a museum and I'll fill it. (Picasso) Give me a forum ...
Join Date: Mar 2003
Posts: 10,537
I wondered what the hell happened to my post! I was interested in seeing what his response might be to suggesting he needed to do a little reading befor pestering Bernstein.
That thread had gone way off track and what I really want is "where do you find the correlation data". Nobody so far knows how to find it.
As far a pestering Bernstein, he has a website with an email address. How can a question like "Mr. Berstein, I've just finished reading The Four Pillars and was wondering where to access correlation data on the web.?" be considered pestering.
I doubt that qualifies as pestering and the emails were several days apart.
Has anyone seen any data on how correlation between asset classes has changed over time. It seems that most correlation data looks at the last 70-80 years. If we took a time machine back to 1960 and analyzed the correlation of asset classes, what would those correlations be (assuming we don't have any knowledge of the assets' performance data 1960-current)? What type of assets would we put in our four pillars portfolio?
Give me a museum and I'll fill it. (Picasso) Give me a forum ...
Join Date: Mar 2003
Posts: 10,537
Quote:
Originally Posted by justin
Has anyone seen any data on how correlation between asset classes has changed over time.* It seems that most correlation data looks at the last 70-80 years.* If we took a time machine back to 1960 and analyzed the correlation of asset classes, what would those correlations be (assuming we don't have any knowledge of the assets' performance data 1960-current)?* What type of assets would we put in our four pillars portfolio?*
I've seen it in finance textbooks. Yes, the correlations change over time.
You have identified the biggest hole in efficient frontier stuff. Correlations and volatilities change all the time. Getting an "optimal" portfolio is an exercise in self delusion.
Having said that, the basic thrust of the research is still of use. If I know that two asset classes are correlated less than 1, it makes sense to include at least some of both of them in my portfolio. How much of each? Well, either you get a mean-variance optimizer to give you a mathematically correct (but practically useless) answer to the fourth decomal point, or you wing it with a rough approximation. Will it be perfect? Of course not. Will it be better than having just one asset class? Almost certainly.
Well, I just reposted a link somebody else had posted in the thread you deleted.* *I consider wanton thread deletion to be pretty rude.* *I can only speak for myself, but I bet many of the people who posted in the thread didn't just post for you to read.* *They probably thought their post would contain information that would be useful or entertaining to other readers (or, like me, they simply post to read their own words).* * Please consider that before you delete another thread that you happened to start.
Bylo has some historical correlation data [doesn't go back pre 1970] for Canadians and Americans. If you want further back than that you'll probably have to go to SBBI.
I think it is Chapter 9 of Roger Gibson's Asset Allocation: Balancing Financial Risk that discusses problems with optimization when assumptions [rates or return, asset correlations, etc] change [with examples]. I found it at my local library.
Great links, Alec. I wonder if anyone has statistically shown what the best "size" of correlation is useful. In my equation below, X. Something like:
10. Re-Check Correlations for the past trailing X years.
20. Reallocate based on these correlations (either every X years or how ever often you choose to rebalance)
30 goto 10
Of course, historical data isnt an accurate measure of yadda yadda. But still.... It would be interesting to see it backtested none-the-less.
Give me a museum and I'll fill it. (Picasso) Give me a forum ...
Join Date: May 2005
Posts: 6,438
raddr has done quite a bit of correlation study; anyone post over there?
__________________ Have Funds, Will Retire "...but do feel free to assert your duly noted opinion on this subject again without benefit of reference or provision of additional information..."
Abstract:* * *
In this paper, we wish to evaluate the performance of simple asset-allocation strategies such as allocating 1/N to each of the N assets available. To do this, we compare the out-of-sample performance of such simple allocation rules to about ten models of optimal asset-allocation (including both static and dynamic models) for ten data sets. We find that the simple assetallocation rule of 1/N is not very inefficient. In fact, it performs quite well out-of-sample: it typically has a higher Sharpe ratio, a higher certainty equivalent value, and a lower turnover than the policies from the optimal asset allocation. The intuition for the good performance of the 1/N policy is that the loss from naive rather than optimal diversification is smaller than the loss arising from having to optimize using moments that have been estimated with error. Simulations show that the performance of policies from optimizing models relative to the 1/N rule improves with the length of the estimation window (which reduces estimation error) and also with N (which increases the gains from optimal diversification). But, even with an estimation window of 50 years, the difference in the performance of the 1/N policy and the policies from models of optimal asset allocation is not statistically significant.
Quote:
Linear Mode
