Article on portfolio management

[Izverg04]

I plan to start investing seriously soon, and I'll start with a passive index-fund approach. I've been reading for a couple of weeks now, posts on Vanguard Diehard forums, articles on efficientfrontier.com, and whatever else I was finding. I think I am an expert on risk management from gambling stuff, so I was looking for something a little more advanced than generic 80% stock / 20% bond recommendations. For the first time, I found almost exactly what I was looking for:

http://hec.osu.edu/people/shanna/chen.pdf

They optimize the expected utility of a portfolio, given different scenarios. The utility function they use, U(W)=W^(1-x)/(1-x) is the same one used in Kelly formalism. What they call the relative risk aversion factor x is exactly the same thing as the Kelly factor (x=4 means Kelly/4 bettor).

There is lots of cool stuff in there. For example, they cite that for a typical American x=6. This is interesting to me, because I think x is higher than that even for professional gamblers on this board.

Still, their results indicate, that any way I slice it, I should be investing 100% of my portfolio in small-cap stocks for a long while. Other results are posted in Table 1 and Figs 1 and 2. What they mean by "Financial Assets" is what you have to invest, and by "Total Wealth" they mean the added value of everything you have now plus what you'll save in the future, discounted by time. Kind of playing bankroll vs. total bankroll.

So does anyone find their analysis unsound? Should I believe their numbers and invest 100% in small-cap? Or have things changed since 1997? I'd also appreciate links to other articles with similar level of discussion.

[DesertCat]

I don't have time to read the article, but it sounds consistant with my personal philosophy. I'm not positive small caps will continue to have an edge over larger cap stocks, but it's not risky to make that bet.

The key to this approach is that you really need to be committed to a long term investment. If you freak out the year when small cap index funds drop 30% and switch to a different approach, you'll kill your performance. You can't put money in an approach like this if you will need it in a couple years (to buy a house, start a business, etc), because you can't predict volatility. This only works for committed retirement funds when you are at least 10+ years (and probably 20+) away from retirement.

If you think you'll need the money sooner, having 20% in a bond fund is a nice

[jively]

I agree with DesertCat that you definitely need the discipline to stick with your allocation. In fact everyone needs that.

Second, consider 3 allocations.

Allocation A:
20% US Large Cap
20% US Large Cap Value
10% US Micro Cap
10% US Small Cap Value
10% REITs
20% International (small and value tilted if possible)
10% Emerging Markets (small and value tilted if possible)

Allocation B:
12% US Large Cap
12% US Large Cap Value
20% US Micro Cap
20% US Small Cap Value
6% REITs
18% International (small and value tilted if possible)
12% Emerging Markets (small and value tilted if possible)

Allocation C:
100% US Small Cap

Allocation A is a globally diversified portfolio that is small cap and value tilted. It should have a very good average return, and the standard deviation is pretty low for the aggressiveness.

Allocation B is more small cap and value tilted, and has more emerging markets. It should have a higher return than Allocation A, and is still globally diversified, so the standard deviation, although higher than Allocation A, is not *that* much higher.

Allocation C is what you are looking at. It is undiversified. It *might* have a slightly higher expected return than Allocation B, but the standard deviation is much higher.

Put another way, both Allocation A and Allocation B should be near the efficient frontier, and Allocation C is not.

I didn't read the article either (I may look at it later), but I like Allocation B a lot more than Allocation C, even if it gives up a slight average return. You are much less likely to have a -30% annual return with Allocation B than with Allocation C, so you are more likely to stick with it in a bad year. And as DesertCat said, sticking with the allocation is extremely important for long-term investment results.

-Tom

P.S. I have a high risk tolerance, and am going to use Allocation B for my own money when I rebalance next month.

[gull]

That is a very interesting paper.

Don't take the advice of 100% small cap literally. For simplicity's sake (I assume), the paper only considered four investment options: small cap, large cap, long-term corporate bonds, and intermediate government bonds. The point is that young people who have long investment horizons can afford to invest very aggressively. People who can afford to take higher risk should definitely invest more in small cap, value, and emerging markets and less in large cap, medium cap, growth, commodity futures, bonds, and cash. The paper only said 100% small cap because the paper used simplified options.

[Izverg04]

[ QUOTE ]
The paper only said 100% small cap because the paper used simplified options.

[/ QUOTE ]
That's true. I'll keep looking for a more comprehensive analysis, but this paper gives very useful insights.