In the last blog entry I referenced a recent article in Geoff Considine’s blog, Portfolioist. This “QPP Analysis” post collects those asset classes and rebuilds them into a portfolio that is analyzed below using the software, Quantext Portfolio Planner (QPP). Platinum members are familiar with the analysis that comes from this software. If not, select the QPP option found in the right sidebar under QPP Analysis and read through some of that information.
Assumption: Readers need to know that I am still assuming the S&P 500 will grow at a rate of 7% over the next six to twelve months. I have my doubts about this percentage gain as my gut tells me it will be somewhere between 5% and 6%. Nevertheless, for this analysis I am using the 7% value.
QPP Analysis: Since Considine did not set up percentage allocations in his article, I was on my own to make those assumptions. The strange starting date of December 19, 2007 was the launch date for EMB. This avoids the problem of the short-term data. All other ETFs have been operational more than five years. I also included VEU and VEA even though Considine does not include them in his article. In the second slide I will show readers why Considine omits them.
Look over the percentages I allocated to each ETF. Not all were assigned a percentage. This portfolio shows a projected return of 7.3% or just a tad above that projected for the S&P 500. Generally, I prefer more of a cushion to the upside as I want the portfolio to perform above that of the S&P 500. The risk or standard deviation is projected to be 12.3%. This is well below our goal of 15%. The “low” SD makes this an appealing portfolio.
Moving down the slide we see the Diversification Metric (DM) is 40% or right on our goal for the current market. The Portfolio Autocorrelation, the least important metric, is acceptable.
Correlation Matrix: Check the high correlation between VTI, VEU, and VEA and you will see why Considine drops the developed international markets asset class. VWO and RWX do not fare much better as they are also highly correlated with VTI. These correlation figures are exactly what Considine is getting at when he says one can strip a portfolio down to about ten asset classes. He mentions that is requires some effort to keep this many asset classes within their target ranges. This is not a problem for users of the TLH Spreadsheet as there is a built-in table on the Dashboard that does all the rebalancing calculations for the user.
This correlation table helps to explain why Considine included VDE and IDU in his asset class mix. If any readers have a combination of ETFs you wish analyzed, send the tickers and the percentages you desire.
Click to enlarge