Another portfolio Fund question

[savory]

I have been working with Portfolio Visualizer and have developed 3 portfolios. I know which of the 3 portfolios have the best performance based on back testing. But, I want learn your thoughts about the selected portfolios in terms of funds and allocation. You will see that it is only the equity part of a 70/30 portfolio. And, I think all are low cost funds. The funds will be in taxable accounts.

Portfolio 1

VTSMX Vanguard Total Stock Mkt Idx Inv - 62%
VGTSX Vanguard Total Intl Stock Index Inv -38%

Portfolio 2

VTSMX Vanguard Total Stock Mkt Idx Inv 70%
FSPSX Fidelity International Index 20%
FSRNX Fidelity Real Estate Index 10%

Portfolio 3

VGTSX Vanguard Total Intl Stock Index Inv - 10%
FSRNX Fidelity Real Estate Index 10%
VTI Vanguard Total Stock Market ETF 70%
FIVFX Fidelity International Capital Apprec 10%

Portfolio Visualizer Results


Portfolio Initial Balance Final Balance CAGR Stdev Best Year

1 $10,000 $22,821 11.11% 11.23% 26.39%

2 $10,000 $24,868 12.33% 10.78% 27.86%

3 $10,000 $25,625 12.76% 10.83% 27.22%

Worst Year Max. Drawdown Sharpe Ratio Sortino Ratio

1 -8.75% -13.44% 0.94 1.48

2 -6.80% -13.20% 1.08 1.72

3 -6.80% -13.25% 1.11 1.78

The last stat is US Market Correction which is .97, .99. ,98. The backtest is 2012-2019

Thanks

 

[OldShooter]

During your test period, the US market has outperformed international stocks. Your test portfolios bear this out, with more home country bias producing better results. For roughly the prior decade the situation was reversed with international winning.

We are simply in VTWAX, buying the world's stocks including the US and not trying to guess a better bias. Those who argue reversion to the mean would say that a higher percentage of non-US stocks will be desirable going forward. So we'll be happy if/when we see that.

Here is a link to a video on home country bias: https://famafrench.dimensional.com/videos/home-bias.aspx

IOW, back testing tells you nothing about the future.

 

[savory]

We are simply in VTWAX, buying the world's stocks including the US and not trying to guess a better bias. Those who argue reversion to the mean would say that a higher percentage of non-US stocks will be desirable going forward. So we'll be happy if/when we see that.

IOW, back testing tells you nothing about the future.

Not to be argumentative but when do you 'believe/know' the market has changed? For example, (and I am going went out on a limb here in terms of my knowledge), in my brief training and exposure to statistical analysis, 3 consistent points outside of the 95% confidence level suggests there has been a significant event that has caused a significant change. It could be temporary or permanent based upon the analysis.

It feels OldShooter as if that type of analysis does not exist here. But there seems like there should be a time when there is a significant enough information that the balance in the investment should tilt. I believe in the buy/hold strategy but I wonder if there is ever a time it should be balanced differently, even ever so slightly. And with that, you might say to me that you do not believe in a buy/hold strategy.

 

[pb4uski]

I'm sure that over the next 20 years, either 1, 2 or 3 will do better than the others.... but I'm not totally sure which one... however, they will likely be close enough to each other that it doesn't really matter.

The OP reminds me of the old saying.... "measure with a micrometer, mark with chalk, cut with an axe".

 

[triangle]

I thought #3 looked a tiny bit odd as it was the only one that has an active component (FIVFX?) along with the passive indexes.

The difference between #1 and #2 will be directly related to how international markets perform relative to US + REITs. REITs have had a great run over the past few decades and many predict that will continue, but who knows. In theory international economies should see higher growth rates than USA but the rule of law is not as strong in many places which leads to lower returns...making it difficult to predict between #1 and #2. Many multinationals are still based in USA so there is some exposure to international markets that way.

I do think 38% in VGTSX is aggressive for someone who lives and retires in the USA, since that slice of investment will grow based on economies/inflation that are different compared to where they live.

 

[DrRoy]

#1 is the least diversified, had the weakest best up year, and the worst down year.

 

[OldShooter]

Not to be argumentative

No worries. I enjoy this type of discussion.

...but when do you 'believe/know' the market has changed? For example, (and I am going went out on a limb here in terms of my knowledge), in my brief training and exposure to statistical analysis, 3 consistent points outside of the 95% confidence level suggests there has been a significant event that has caused a significant change. It could be temporary or permanent based upon the analysis.

It is very tempting to try to play mathematical games with stock price series. I know. My first attempts at investing (1972) involved trying to use Fourier transforms to predict prices. It didn't work.

I don't believe the concept of "confidence level" can really exist for stock prices, because all that math is based on Gaussian distributions. Investments are not Gaussian.

First, the distributions have "fat tails" and the left (negative) tail is bigger than the right tail. Second, the distribution is not centered on zero; it moves to the right by a few percentage points per year. Third, all the nice math you get with Gaussians is based on the idea that the samples are independent. The fact that momentum (correlated samples) exists in stock prices is pretty much settled science even though it is almost impossible for traders to use it to make money.

We get proof that a Gaussian model doesn't work every once in a while when we get a five-sigma or more event like 2007/8. (Maybe it was seven sigma; I don't remember.)

It feels OldShooter as if that type of analysis does not exist here.

Correct because, sadly, it doesn't work.

But there seems like there should be a time when there is a significant enough information that the balance in the investment should tilt.

Well, the other thing about the market is that it is self-correcting for mispricings. For example, if a stock always went down on Friday afternoon and rose on Monday the obvious strategy would be to buy on Friday and sell on Monday. But there are tens of thousand (maybe hundreds of thousands) of people scrutinizing the market for mispricings, so very quickly the Friday prices would be bid up and the Monday prices would become depressed. Game over. Just looking at public mutual funds there are about 10,000 and most of them are chasing the 3600 or so stocks in the US domestic market. Very hard for any mispricing to go undetected.

I believe in the buy/hold strategy but I wonder if there is ever a time it should be balanced differently, even ever so slightly.

Absolutely there are those times and they are easily visible in the rear view mirror. Through the windshield, not so much. This is the trap when backtesting; it is easy to follow Will Rogers' advice: "If it don't go up, don't buy it."

IAnd with that, you might say to me that you do not believe in a buy/hold strategy.

Au contraire, I believe. The reason that buy and hold wins is that internally goofy stock price distribution has (so far/100 years or so) always trended toward the right. So us B&H guys don't have to worry about all the craziness that might be going on day to day. We just ride the trend. (DW and I look at our portfolio once a year and some years we even make a trade.)

I think you would enjoy a couple of books by Nassim Taleb: "Fooled by Randomness" and "The Black Swan." He can be kind of a lunatic but he can also be very insightful and is a fun read.

 

[OldShooter]

#1 is the least diversified, had the weakest best up year, and the worst down year.

Actually its the opposite -- #1 is the most diversified, although the reduction in diversification in #2 and #3 is negligible.

Here's how that happens: VTSMX has the whole US market including the REITS. Adding a REIT fund reduces diversification by largely or totally duplicating holdings that are already in VTSMX. Sometimes this is referred to as a "tilt." No big deal, but the essence of any tilt is to reduce diversification by adding specific concentration(s). Small cap and value tilts are very popular too. Done with broader brushes and more paint, you get factor-based investing.

 

[savory]

I think you would enjoy a couple of books by Nassim Taleb: "Fooled by Randomness" and "The Black Swan." He can be kind of a lunatic but he can also be very insightful and is a fun read.

Thanks for your thoughtful reply and your follow=up to Dr Roy's comment. I will look for these books.