It's not to my taste and I seriously doubt that it's much better than two-thirds total stock, one third total bond, but I don't think there's anything seriously wrong with it...
...except the question of whether you have the conviction to stick to it.
I think one of the most important things long-term investors need to do is let go of the idea of finding the optimum portfolio. You must accept that you can't know what it is, settle for "good enough," and stick to it. During the times when it underperforms, and there will always be such times, you say "that's OK, it's good enough."
Now, compared to two-thirds total stock, one third total bond, what you've done is mostly to increase your risk slightly in hope of increasing reward. You haven't increased it a whole lot, it's not insanely risky, but it's riskier. I did a backtest. I used VIGRX for large growth, VSVIX for small value, and VWEHX for junk bonds; VIGRX=VUG, VSVIX=VBR, but they have longer histories.
Source
Blue, 34% VIGRX, 33% VISVX, 33% VWEHX
Red, 67% VTSMX, 33% VBMFX
![Image]()
over that time period, it definitely had higher return, almost a full 1%/year more. But the reason for it was not so much any fine-tuned factor adjustment, it's just that you upped the risk. Generally speaking, a good way to kid yourself is to compare portfolio returns without even trying to compare the risk somehow. Because returns are very visible, but risk isn't.
The numbers to look at here are the Sharpe and Sortino ratios, which are measures of risk-adjusted return. Units of return per unit of risk. (The Sharpe ratio "cares" about uncertainty in either direction, the Sortino ratio only "cares" about downside risk). In both cases, the simple two-thirds total stock, one third total bond portfolio had very slightly higher Sharpe and Sortino ratios. The difference isn't enough to mean anything, but it's enough to show that the higher return of your portfolio merely commensurate with the higher risk.
But if the higher return is attributable to higher risk, then you need to observe that this isn't the only way to get a little more risk. Is it a specially good way? Does it give clearly better result than other ways, e.g. just upping the stock allocation?
And the problem with "just taking a little more risk" comes during the bad spells. If you'd had the bad luck to invest $10,000 at the start of 2020, you'd have seen $7,000 in your account at the end of March, while the simple portfolio would have had $8,000.
Much more seriously, because of lasting much longer, during 2008-2009, your portfolio would have declined -45% while the simple portfolio declined -38%. The higher return over eighteen years was associated with deeper dips at the times when the risk showed up.
Could you have stayed the course, or would you have said "this isn't working, I'd better make a change?"
In the course of this discussion, you seem to have changed your ideas, from specifying junk bonds to specifying either junk bonds or long-term Treasury bonds. If you actually adopt either portfolio, will you be able to stick to it, or will you find yourself constantly tinkering as you learn of new ideas and experience new market conditions?
...except the question of whether you have the conviction to stick to it.
I think one of the most important things long-term investors need to do is let go of the idea of finding the optimum portfolio. You must accept that you can't know what it is, settle for "good enough," and stick to it. During the times when it underperforms, and there will always be such times, you say "that's OK, it's good enough."
Now, compared to two-thirds total stock, one third total bond, what you've done is mostly to increase your risk slightly in hope of increasing reward. You haven't increased it a whole lot, it's not insanely risky, but it's riskier. I did a backtest. I used VIGRX for large growth, VSVIX for small value, and VWEHX for junk bonds; VIGRX=VUG, VSVIX=VBR, but they have longer histories.
Source
Blue, 34% VIGRX, 33% VISVX, 33% VWEHX
Red, 67% VTSMX, 33% VBMFX
over that time period, it definitely had higher return, almost a full 1%/year more. But the reason for it was not so much any fine-tuned factor adjustment, it's just that you upped the risk. Generally speaking, a good way to kid yourself is to compare portfolio returns without even trying to compare the risk somehow. Because returns are very visible, but risk isn't.
The numbers to look at here are the Sharpe and Sortino ratios, which are measures of risk-adjusted return. Units of return per unit of risk. (The Sharpe ratio "cares" about uncertainty in either direction, the Sortino ratio only "cares" about downside risk). In both cases, the simple two-thirds total stock, one third total bond portfolio had very slightly higher Sharpe and Sortino ratios. The difference isn't enough to mean anything, but it's enough to show that the higher return of your portfolio merely commensurate with the higher risk.
But if the higher return is attributable to higher risk, then you need to observe that this isn't the only way to get a little more risk. Is it a specially good way? Does it give clearly better result than other ways, e.g. just upping the stock allocation?
And the problem with "just taking a little more risk" comes during the bad spells. If you'd had the bad luck to invest $10,000 at the start of 2020, you'd have seen $7,000 in your account at the end of March, while the simple portfolio would have had $8,000.
Much more seriously, because of lasting much longer, during 2008-2009, your portfolio would have declined -45% while the simple portfolio declined -38%. The higher return over eighteen years was associated with deeper dips at the times when the risk showed up.
Could you have stayed the course, or would you have said "this isn't working, I'd better make a change?"
In the course of this discussion, you seem to have changed your ideas, from specifying junk bonds to specifying either junk bonds or long-term Treasury bonds. If you actually adopt either portfolio, will you be able to stick to it, or will you find yourself constantly tinkering as you learn of new ideas and experience new market conditions?
Statistics: Posted by nisiprius — Sat Nov 30, 2024 8:21 am — Replies 28 — Views 2057