Bohr Portfolio Update

If you are looking for a portfolio that needs attention, look no further than the Dashboard of the Bohr Portfolio as presented below.  Asset classes coded with the purplish color are all under target while the red are above target.  To rectify these issues and bring the portfolio back into balance, I have a number of limit orders in place.  As the market continues to rise, the current prices are walking away from my limit order prices.  I don't plan to chase this market as I expect a correction sometime this summer.  Therefore, I am sticking with my limit orders.  So far my stubbornness or patience is costing money.

Available cash was a constraint in bringing every asset class back into balance unless I sold off some bond and income holdings.  I plan to do that next week.  With TLT near $94.00, it is prudent to sell some shares.  In addition, I was just told that this particular portfolio will receive an infusion of new cash.  That is always good news as available cash makes rebalancing easier.  However, I plan to be slow in putting new cash to work.  One exception I will likely make has to do with an attractive stock that was number one on the "Creme List" this week.  The Bohr currently holds one stock, Procter and Gamble (PG).  I would like to increase the number of stocks to increase diversity.

It is reasonable to ask, how can one improve diversity be adding stocks to a portfolio already holding hundreds of stocks within a wide variety of ETFs?  This is where I rely on results from the Quantext Portfolio Planning (QPP) analysis.  Long-time readers have seen the benefit of adding a few highly selected stocks to an ETF oriented portfolio.  It is important to select stocks that have a low correlation with the ETF oriented portfolio.

Below are the performance numbers and asset make-up of the Bohr Portfolio.  Note the very high Sortino Ratio (SR) and Retirement Ratio (RR) values.  In reality, the SR value is even higher, but I needed to insert a "fudge factor" since the portfolio has always outperformed the VTSMX benchmark.  Without the "fudge factor" the semi-variance denominator becomes zero and the ratio blows up into infinity.  To fully understand what is happening, the reader needs to dig into the calculations in the SR worksheet of the TLH Spreadsheet.

[Read more...]

Portfolio Performance Data – 29 April 2011

While this was a good week for all portfolios in absolute terms, relatively speaking, nearly every portfolio lost ground to one or more of the benchmarks.  Advances from last week were essentially wiped out this week.  When the market is in a strong bull phase as is now the case, it is difficult for a broadly diversified portfolio to keep up with U.S. equities and that is what we saw happen this week.  Bonds and any sort of income oriented ETF are a drag on performance in this type of market.  This will not continue and it will be interesting to see what happens over the summer when we expect to see this market cool off in a major way.

Looking over the entire data table, we are not displeased with the results.  Several portfolios have respectable leads on the VTSMX benchmark and most of the Sortino and Retirement Ratio (RR) values are either positive or within striking distance of moving above zero.  In a few portfolios, we did see advances in the RR value.

Portfolio Performance - 04/29/2011

Portfolio Last Update Launch Date Tracking Tool Port. IRR ITA Index Diff Port. vs. ITA Index VTSMX IRR Diff. Port. vs. VTSMX Index IR SR RR
AA-Mosaic 03/31/2011 07/21/1999 Captool 2.86% NA NA 1.30% 1.56% 0.05 NA NA
Curie 04/29/2011 12/26/2007 TLH SS 6.8% 1.7% 5.1% 3.0% 3.8% NA 11.6 6.4
Mosaic2 03/31/2011 07/19/1999 Captool 5.12% NA NA 1.31% 3.81% 0.15 NA NA
Newton 04/29/2011 06/02/2008 TLH SS 11.0% 14.6% -3.6% 9.6% 1.4% NA 1.0 1.0
Passive Port. 04/29/2011 12/01/2000 Captool 5.83% NA NA 3.35% 2.48% 0.70 NA NA
Schrodinger 04/29/2011 12/01/2000 TLH SS 5.5% 0.5% 0.0% 3.5% 2.0% NA 4.0 1.6
Jane 03/31/2011 02/14/1997 Captool 9.35% NA NA 5.25% 4.1% 0.54 NA NA
Einstein 04/29/2011 06/30/2008 TLH SS 17.0% 20.1% -3.1% 13.3% 3.7% NA 4.9 4.9
Gauss 03/31/2011 02/19/1997 Captool 9.52% NA NA 2.93% 3.25% 0.22 NA NA
Kepler 04/29/2011 11/01/2008 TLH SS 22.4% 25.8% -3.3% 22.1% 0.3% NA 0.17 0.17
Scrappy 03/31/2011 08/14/2008 Captool 13.58% NA NA 6.9% 6.68% NA NA NA
Bohr 04/29/2011 08/14/2008 TLH SS 14.5% 15.4% -0.9% 7.9% 6.6% NA 4.6 4.6
Kenilworth 04/29/2011 08/18/2010 TLH SS 31.2% 35.2% -4.0% 39.8% -8.6% NA -0.06 -0.06
Projects 04/29/2011 12/01/2000 Captool 5.97% NA NA 3.03% 2.94% 1.33 NA NA
Washington 03/31/2011 06/18/1999 Captool 3.18% NA NA -0.18% 3.36% 0.28 NA NA
Maxwell 04/29/2011 12/25/2000 TLH SS 1.1% 11.5% -10.4% 4.2% -4.1% NA -0.18 -0.18
Adams 03/31/2011 06/18/1999 Captool 3.55% NA NA -0.18% 3.73% 0.71 NA NA
Euclid 04/29/2011 06/30/1999 TLH SS 2.1% 10.6% -8.5% 5.4% -3.3% NA -0.15 -0.15
Jefferson 03/31/2011 03/13/2008 Captool 6.11% NA NA -3.05% 9.16% NA NA NA
Madison 04/29/2011 03/13/2008 TLH SS 6.8% 41.9% -35.1% 12.3% -5.4% NA -0.23 -0.23

[Read more...]

You Can’t Manage What You Don’t Measure

Proper portfolio management requires a number of measurement activities.  How is an investor to know what is going on within their portfolio if they do not take the time to accurately measure not only the performance of the portfolio, but the performance of an appropriate benchmark, and portfolio uncertainty or risk?  Note the emphasis on appropriate.  Managing a single portfolio, as is the case with most investors, requires accurate measurement of the Internal Rate of Return (IRR).  This is a starting point.  There are numerous software programs available that perform this task quite well.  Each year, the American Association of Individual Investors (AAII) reviews portfolio tracking software programs.  One highly ranked program is produced and supported by Quant IX and is titled, "Investment Account Manager 2.0."  I've used this software and can vouch for its accuracy and ease of use. As a stand-alone program, it is one of the best available.

Another program I use is Captool.  Unfortunately, the "poor person" version of this software is no longer supported as Captools migrated to serving larger investment firms. A third, and affordable portfolio measurement tool, is an on-line service at http://www.bivio.com.  While designed for investment clubs, there is no reason an individual investor could not use this software to monitor their portfolio.  There is an annual cost for using their services.  Bivio has the advantage of providing IRR measurements for a number of benchmarks.  I've use Bivio and found it an extremely useful portfolio tracking tool.

While I use an old version of Captool to manage ten of the twenty portfolios tracked on this blog, my preference of portfolio tracking software tools is the TLH Spreadsheet.  A sample spreadsheet is available and I am working on an updated version.  The new version should be ready for release sometime in May.

Before going into some of the highlights of the TLH Spreadsheet, here are the benchmark guidelines used. Nancy L. Jacob wrote an essay on 'Evaluating Investment Performance' and it is included in Peter L. Bernstein's outstanding collection, "The Portable MBA in Investment."  Quoting from Jacob's article, a good performance benchmark is:

  • Objectively constructed.
  • Investable.
  • Specifiable in advance.

A good performance benchmark has:

  • Easily observed performance.
  • Unambiguous composition.
  • Same style and risk as manager.

The TLH Spreadsheet uses these guidelines to create and calculate benchmarks.  Here are the primary strengths of the TLH Spreadsheet.

  • The TLH provides an accurate measurement for the Internal Rate of Return (IRR) of the portfolio.
  • Within the spreadsheet is the Dashboard worksheet.  This worksheet is where the user lays out the Strategic Asset Allocation plan for the portfolio.
  • Rebalancing help is available through a customized data table.  This is found within the Dashboard worksheet.
  • Should investors build their portfolio around sectors instead of asset classes, graphical representations are available so the investor can see how well their portfolio matches either the S&P 500 sectors or a total market index sector allocation.
  • The TLH tracks the VTSMX, VFINX, and VGTSX index benchmarks.
  • A customized benchmark, ITA Index, is available.  While not perfect, the ITA Index does a good job of tracking the portfolio performance if multiple asset classes are used.
  • Uncertainty or risk measurements are available.  Of particular interest is the Retirement Ratio

What are the negatives to using the TLH Spreadsheet as a portfolio measurement tool?  The biggest disadvantage is the learning curve required to use all parts of the spreadsheet.  To lower the learning hurdles I've supplied a number of audio/video sessions and these are available to Platinum members.  More are added when needed or requested by users.

Measure and manage.  If one does not adequately measure portfolio performance, uncertainty, and asset allocation percentages, it is nearly impossible to adequately manage a portfolio.

Is It Worth 50 Basis Points? Part III

Example three in this series uses the last three years of data for the return and volatility projections.  As in the last two blog posts, not much changes. The projected Return/Uncertainty ratio remains at 0.49.  Part of the reason for little change is that 40% of this portfolio is made up of bond and income assets and they tend to dominate the portfolio.  Tamping down volatility is definitely a dominate factor in the asset makeup of this portfolio.

The question remains, is it possible to construct a portfolio using non-managed ETFs, avoid the management fees, and come up with a portfolio that provides a higher projected return, holds down volatility, and increases the diversity as measured by the Diversification Metric?  What is it worth to avoid the responsibility of portfolio management is another question?

ITA Wealth Management readers who took my recommendation and read Goldie and Murray's book, "The Investment Answer" will remember decision #1 is 'The Do-It-Yourself' decision.  Goldie and Murray recommend the investor seek out a fee based advisor.  I take the opposite position.  My recommendation is to do it yourself and save the fees as they are not trivial for large portfolios.  Each investor needs to make this decision. 

[Read more...]

Is It Worth 50 Basis Points? Part II

Yesterday, I posted a conservative portfolio built around index funds only available through qualified advisors.  The portfolio projections were based on the last five years of data.  The following screen shot presents future projections of the same portfolio using the last four years of data.  As readers will see, nothing really changes in the way of projections.  What does change is the actual performance return and volatility range over the four-year period.  If you compare the performance of the portfolio with the performance of the S&P 500, note that the S&P 500 does not include dividends.  This would add nearly 2% points to the S&P value of -.28%.  Also keep in mind that the S&P 500 is not an appropriate benchmark for this portfolio as the portfolio includes a high percentage of investments outside the large-cap asset classes.

The question still remains, does it make sense to pay 50 basis points or higher to gain access to these funds and professional advice?  The answer is not clear cut and will certainly vary depending on how much the investor wishes to become involved in portfolio management.

[Read more...]

Is It Worth 50 Basis Points – Or Higher?

Approached recently about the possibility of investing in index funds that are only available through fee based advisors, I decided to take a closer look.  The question of whether to pay 50 to 100 basis points for access to specialized index funds is an honest question.  The first step was to define what risk (portfolio uncertainty) might be required as this would determine the percentage to allocate to different index funds.  That "risk test" resulted in a rather conservative portfolio and the allocations are available in the screen shots below.  I hope to run the numbers on a higher uncertainty portfolio in a later post.

Disclaimer and Assumptions:  Readers need to keep in mind that the following analysis comes from using the Quantext Portfolio Planner (QPP).  While it is a sophisticated tool, it is still limited by the dangers inherent in extrapolation of data.  Further, past historical data has seen extreme volatility so I decided to run three sets of data using different time frames.  These will come in different posts so pay attention to the time spans for each analysis.  In the first screen shot I am using the last five years of data.

The S&P 500 is assumed to grow at 7% over the next year.  This value is below the default setting and it will negatively impact the projected return.  Nevertheless, one can compare the projected return of the portfolio with the S&P 500 projection.

Examine this data and compare it to many of the Strategic Asset Allocation plans presented throughout ITA Wealth Management.

The projected return is 5.9% (rounded) or about one percentage point below what is expected for the S&P 500.  Keep in mind that one of our benchmarks consistently outperforms the S&P 500, so 5.9% is not a robust return.  To be fair, the projected uncertainty is a low 11.9%.  These values result in a Return/Uncertainty ratio of 0.49, or well below our goal of 0.60.

The Diversification Metric (DM) is nothing to write home about as it is a modest 30%, or 10% points below our goal.

Here we have the Correlation Matrix.  Once more, we find diversification when we load up on those bond and income index funds.

It is no mystery this portfolio is constructed from DFA index funds.  These funds are only available through certified advisors.  Even with the potential errors that arise from making future projections using QPP, I continue to be reluctant to pay an advisor for something the average person can do on their own.  Does it make sense to pay an annual fee of 50 basis points for a 0.49 Return/Uncertainty ratio and a DM of 30%?

MACD Defined

Yesterday, I wrote a piece on "Sell in May and Go Away," Sy Harding's Seasonal Timing Strategy (STS).  Built into that strategy is something called MACD.  MACD is an acronym for "Moving Average Convergence-Divergence" Trading Method.  I first read about MACD back in 1979 when Gerald Appel published a pamphlet under the same title. In that 22 page pamphlet he laid out the fundamentals for MACD technical analysis.  One year later, in 1980, Appel co-authored "Stock Market Trading Systems" with Fred Hitschler.  That hardbound book expanded on the material published in the pamphlet.

Exactly how does one calculate the MACD?

1. Determine a Smoothing Constant.  Before one can compute an exponential average, one must derive a smoothing constant.  Assume you are interested in coming up with a 50-Day EMA.  

Smoothing Constant = 2/(50 + 1) = 2/51 = 0.0392 or rounded to 0.04

2. Apply the Smoothing Constant to come up with the first Exponential Moving Average (EMA).

New EMA = Smoothing Constant * (Today's Price Close – Yesterday's Exponential Moving Average) + Yesterday's EMA

To start this calculation, one needs to pick a price for Yesterday's Exponential Moving Average and it takes the number of days, 50 in this example, to develop an accurate EMA.

Example:  Assume Yesterday's EMA was $50.5 and the ETF closed at $51.2 today.  The new 50-Day EMA would look like the following.

New EMA = 0.0392 * ($51.2 – $50.5) + $50.5 = $50.53   Notice the slight increase that gives rise to the new Exponential Moving Average.

These are just the beginning steps in calculating the MACD.  We first came up with a smoothing constant based on the number of days we want to use for our first EMA calculation.  Then we run the EMA calculation as explained above.  A very common EMA is the 200-Day EMA.

Back in 1980, MACD was called MMACDTS for Multiple Moving Average Convergence-Divergence Trading System.  We are fortunate that did not stick and we are relegated to MACD. 

3.  The final step in coming up with MACD is to calculate three different EMAs. MACD is the difference between two Exponential Moving Averages and that value is compared to a trigger or signal EMA, normally a 9-Day EMA.  Are you lost?  I hope not.

I capture my MACD information from StockCharts and the default MACD setting is the difference between the 13-Day EMA and 26-Day EMA.

Check the Wikipedia reference and you will see the blue line is the difference between the 13- and 26-Day EMAs.  That line is then compared with the signal line or the 9-Day EMA.  When the blue line crosses from below to above the red line, a buy signal is generated.  When the blue line crosses from above to below the red line, a sell signal is generated.

If you check StockCharts for nearly any investment, you will observe this involves a lot of trading and that is a major reason for not using MACD.  Harding gets around this problem by applying the MACD only twice a year.  Once in the fall and then again in the spring.  While Harding's Seasonal Timing Strategy worked during the first decade of the 2000s, its long-term record is less than stellar.

There you have the basics from first establishing a smoothing constant to calculating one Exponential Moving Average.  Three EMAs are needed to calculate the MACD.  In the 1980s all these calculations were handled inside a spreadsheet known as Visicalc.  That is no longer necessary as StockCharts does all the work for us.

“New Normal” Portfolio Analysis

Nearly three months passed since I last ran an analysis on the "New Normal" portfolio. At that time the projected return was 7.7% when I assumed the S&P 500 would return 7.0% over the next year.  The projected uncertainty was a modest 13.1% giving a Return/Uncertainty ratio of 0.58.  This portfolio is well diversified and it shows with a Diversification Metric (DM) of 46%.  Those were the figures back in February of 2011.

This morning I ran the numbers again and there is not much change.  As one can see from the screen shot below, the projected return is a tad higher and we see an increase in projected uncertainty.  The Return/Uncertainty ratio does not change and the DM drops one percentage point.

I will need to check with the developer of this portfolio to make sure the stocks are still the same.  This particular portfolio holds ABT, AFL, PG, TEVA, and JNJ.  All are well-known and solid companies.

Readers generally find the Correlation Matrix of interest and I publish it below.  Note the many blue background investments dotted throughout this portfolio.  This is why the Diversification Metric is above 40%.

This particular portfolio does better than the market in down markets, but tends to lag slightly in up markets.  It is definitely more defensive in nature than the seven portfolios Platinum readers have access to on this blog.

Basic Portfolio Plus

Building on the five basic asset classes, I made a few substitutions and added more asset classes.  Platinum members can see this portfolio enhanced the Basic Five.  I show both the correlation matrix and the projected return and uncertainty values.  A few of the assumptions were altered as articulated below.

[Read more...]

Basic Five Asset Class Portfolio

What are the basic asset classes that belong in every portfolio?  Investors beginning the portfolio construction process would do well to include each of the following five asset classes in their portfolio.  The ETF ticker suggestions are among those I use to populate these five asset classes.  This group of ETFs provide broad coverage of the world market, although there are still a few gaps to close.
 

  • United States Equities   (VTI)
  • International Markets (Developed)   (VEU)
  • Emerging Markets   (VWO)
  • REITs   (VNQ)
  • Bonds   (BND)

 

For ease of analysis I allocated 20% to each of the above five asset classes.  As a further assumption, I set the S&P 500 return projection to 7.3% for the next year.  The projected return for this basic portfolio is 8.6% with a rather high uncertainty of 17.2%.  This combination gives a Return/Uncertainty ratio of 0.50.  We prefer to see this value above 0.60. 

The Diversification Metric is only 17%.  Only BND provides a low correlation as the other four ETFs are highly correlated. 

 

Below is what I call the "Delta Index" and it is infrequently posted.  What this data table shows is a probability indicator as to whether or not it is an appropriate time to be adding the ETF to the portfolio.

[Read more...]