Stress Testing the 10 ETF Portfolio

A Seeking Alpha reader showed interest in the 10-ETF Portfolio and asked how such a portfolio might stand up under a stress test as I applied to another portfolio. Before running such an analysis, one needs to understand the assumptions.  For this analysis I assume the investor is 45 years old, has saved $250,000, saves $12,000 per year, and will retire on $50,000 per year upon retirement in 2032 at age 66.  Inflation is assumed to be 3.5%.

The chances of running out of money are not all that great.  While I would like to see that 10% chance begin somewhere in the 80s, I'll settle for the Monte Carlo probability results shown below.  Now we need to see what happens when a three sigma event occurs and we lose nearly 50% of the probability.

Become a Platinum member to see more of the blog.

[Read more...]

Seeking Alpha Questions

In response to a Seeking Alpha article, I was asked the following questions.  Here are the questions and my replies.

1. About the Quantext data box in your article. For the Portfolio Stats in the upper right-hand corner, what's the time frame? I picked 15 years for my comparison with VBINX, but I didn't know what time frame the Portfolio Stats refer to.

Geoff Considine, developer of the QPP spreadsheet recommends using three or four years of data.  In every article he published, he used three years.  That is generally the time frame I use, but I may go back further for this reason.  Using three years of historical data is not getting us into the period when the stock market was sinking like a stone.  Going back four years provides a little different outlook.  Personally, I think it is a good idea to run 3, 4, and 5 year periods to see what difference shows up in the critical projections.  If readers check the data inside the yellow background, the time frame is always stated.

2. Also about the Average Annual Return in the upper right-hand corner: is this simply the weighted average of all the funds in the list, or does Quantex do something else to calculate the average?

I'm not sure if it is a weighted calculation.  I think it is but I would need to check the calculations to be sure.  The actual equations behind the calculations are hidden from the user.

3. I was surprised that both the Diversification Metric and the Portfolio Autocorrelation were better for the 10-ETF Starter Portfolio than for the 25-Fund Retirement Portfolo (30 vs 26 Diversification, 21.27 vs 29.22 Autocorrelation). I would expect the fund with more ETFs, including commodities, to be more diverse and less correlated. Any reason why? Or have I misunderstood the stats?

The Diversification Metric is controlled by the Correlation Metrics worksheet and that worksheet is a workhorse when it comes to calculating how the different assets impact each other and how the combination compares to the S&P 500.  When I run out QPP analysis on portfolios, I'm amazed how highly correlated the different ETFs are to each other.  For example, the U.S. and developed international markets are highly correlated.  Ibbotson and Associates show this in "recent" analysis where our portfolios are primarily driven by the broad markets.  As I recall, Ibbotson concludes that 75% of portfolio movement is tied to the broad markets.

4. Vanguard's web page for VBINX seems to indicate the the fund started trading on 11/09/1992, not 1996. But that's irrelevant to this discussion, since I just chose it as a proxy for my comparison.

I picked up my data from Yahoo Finance.  I don't know the data source Considine uses for QPP.  I suspect it is Yahoo as I don't think he would pay top dollar for an independent source.

5. My interest, as I said in my comment, is to figure out how someone can create an easily managed retirement portfolio (like your 10-ETF portfolio) that has a historical return of 7% (over whatever period is used in the upper right-hand corner of the Quantex box) with a Return/SD greater than .6. It's an interesting question. It essentially involves getting the SD down to 11.66 if the Return is 7%.

I think this is possible, and I am working on coming up with such a portfolio.  However, I want to go one better and that is to see the projected return be one percentage point above that projected for the S&P 500.  That moves the return up to 8%.  I'm not quite there without using a hedge ETF.

6. And for people who can't or are not willing to buy ETFs and conscientiously rebalance their accounts, what Vanguard or Fidelity funds can be used to mimic a relatively diversified portfolio with a long-term 7% return and Return/SD of .6. Over a 15-year timeframe, VBINX seems almost to do that, although it's not very diversified.

One can get the returns up to 7%, but it is difficult to hold down the uncertainty.  I've not tried to come up with such a portfolio using index mutual funds. 

Kepler Portfolio Review

A small amount of cash was deposited into the Kepler and that is one of the reasons for this review.  As usual, we first check the asset allocation worksheet, known as the Dashboard, to see what asset classes are out of balance.  Pay little attention to the Cash asset class as that one is always over target as we try to stay fully invested.

Platinum readers can see that only Mid-Cap Blend and Large-Cap Blend are below target.  Limit orders are in place to bring these two asset classes back into balance.  Paying attention to two other asset classes, note that developed international is above target and emerging markets is below target, even though both are within the 25% threshold limits.  What I did was to set a conditional order to sell 50 shares of VEU at a price above the current price, and when that happens, buy VWO at a price below the current price.  It may be months before this combination takes place, but this is the type of move one makes to try to eke out small differences so as to outperform the benchmarks.  TDAmeritrade permits one to go out four months with a limit orders and I am able to extend that another month.  If it does not happen in a few months, I just go into the TDAmeritrade account and extend the time frame.

Platinum membership is available for $5.00 per month.  Learn how to construct, manage, and monitor your portfolio.

[Read more...]