Does It Make Sense To Include Bonds And Income ETFs In Your Portfolio? [View article]
To the reader who requested EDV, VGLT, and BLV be added to the next analysis, I've done that. VGLT does not have a three-year record so I may eliminate that ticker.
Portfolio Rebalancing: Threshold Limits And Correlation Of Assets Are Determining Factors [View article]
Correlation data simply tells us the dependence between two or more variables. The correlation matrix lays out how different tickers, in the example in this article, are dependent on each other. When it comes to investments, if we want to lower portfolio volatility, we attempt to find investments that will move in different directions in different market conditions.
A good example is the Harry Browne "Permanent Portfolio."
Portfolio Rebalancing: Threshold Limits And Correlation Of Assets Are Determining Factors [View article]
Jreddick,
Good question and I will try to answer as best I can. When faced with a question like this one, I find it useful to set up extremes to help understand what is happening. Here goes. Assume we have an extreme portfolio where 90% of the portfolio is invested in VTI (U.S. Equities) and 10% in BND (broad bond ETF) giving us a stock to bond ratio of 90/10. I ran the analysis moments ago and I see where the correlation between VTI and BND is a very low 8%, or something we might expect. However, in the context of the 90/10 portfolio, VTI has a correlation of 100% whereas BND is 10%. In other words, VTI dominates the portfolio since it controls 90% of the assets.
Now we move to the opposite extreme where only 10% is invested in VTI and 90% is allocated to the bond ETF, BND. While the correlation between VTI and BND remains the same at 8%, that in no way takes into consideration the allocation within the portfolio. In the context of the overloaded exposure to BND, VTI has a 52% correlation and BND an 89% correlation. The construction of the portfolio alters the correlation percentages. Even at 99% allocated to BND and only 1% allocated to VTI, within the portfolio, BND has a 100% correlation and VTI a 13% correlation. These do not need to add to 100%. As the allocation to BND increases, so does the correlation of the overall portfolio. When 90% is allocated to VTI, that one investment essentially controls the movement of the portfolio. When 90% is allocated to BND, it too tends to control the movement of the portfolio, but not quite to the extent as VTI.
While we desire low correlated assets, one only takes advantage of the low correlations if the asset allocation plan devotes a sufficient percentage to the low correlated assets so as to make a difference.
Does It Make Sense To Include Bonds And Income ETFs In Your Portfolio? [View article]
Larry,
No, I don't track any of the tickers mentioned in your reply. In no way does that imply your holdings are deficient.
What I do recommend is an accurate method for calculating the performance of the portfolio and all constituent parts. In addition, I'm a strong advocate for benchmarking the portfolio using an appropriate benchmark. I don't think one can easily compare portfolios as each has a different cash flow and launch dates vary.
How To Construct A Customized Benchmark For Your Portfolio [View article]
Disallusioned,
I both agree and disagree with the first paragraph. In the TLH Spreadsheet I have four benchmarks. 1) VTSMX - as this benchmark will tell me how well I am doing with respect to the total U.S. Equities market. This is similar to you using the Toronto index. 2) VFINX - as this benchmark provides a reference how well the portfolio is performing vs. an investable S&P 500 index. 3) VGTSX or the international market. This index fund provides a reference for how well international investments stack up. This index fund has been lagging the VTSMX by a significant percentage and that helps explain why a portfolio that contains a significant percentage of international securities is bound to lag the U.S. market. 4) The ITA Index or the customized benchmark is the one discussed in the article. This benchmark tells me how well the portfolio is performing compared to my Strategic Asset Allocation plan.
It makes little sense to hold a portfolio of 40% in bonds and then expect the portfolio to outperform the S&P 500 over the long run. Why use the S&P 500 as a benchmark if the portfolio holds a significant percentage in small-cap stocks. While the TLH Spreadsheet provides a comparison with the S&P 500, it provides additional information to the money manager.
As for risk, I use three different measurements. 1) Information Ratio. 2) Sortino Ratio and 3) Retirement Ratio (RR). RR is my own design and a variation of the Sortino Ratio. The SR and RR risk measurements are unique in that they use a semi-variance calculation instead of a mean-variance calculation. The money manager is only penalized with the variation is below the benchmark, not when the volatility is above the benchmark as we want volatility to the upside.
As I mentioned in the article, the ITA Index is not a perfect benchmark, but it goes a long way toward telling the money manager how well they are managing a portfolio based on their Strategic Asset Allocation plan for the portfolio.
The Truth About Active Vs. Passive Investing [View article]
Kurtis,
I just came across your article when I was searching for Active vs. Passive investing.
Question: Is the S&P 500 the benchmark used in every comparison? If the actively managed mutual funds are making use of small-cap stocks why not use a more appropriate benchmark?
Boost Core ETF Portfolio With Low Beta And Piotroski Stocks [View article]
Delta David,
Yes, adding GAI and PFIN is an experiment. I'm using these "Piotroski" stocks with one portfolio. While it is very early in the experiment, here are my IRR results as of moments ago.
CRAI = 16.2% Closed out this position when the stock dropped below 7 on Piotroski's High F-Score. PFIN = 73.1% Exaggerated value due to short time frame. GAI = 8958.2% VERY exaggerated due to short time frame.
I check the Piotroski High F-Score screen every week to see if the stock still holds.
Swensen Portfolio Supercharged With 2 Piotroski Stocks [View article]
Depraved,
You are not missing anything that has come to my attention. Right now GAI has a High-F score and I will continue to hold it in a portfolio so long as it maintains a rating of 7 or higher. I generally wait to buy when the score is 9, or a perfect Piotroski score.
One needs to understand how a program works and then add your own features if the program permits adding improvements. Follow this up by appropriately benchmarking the portfolio to see if one is adding alpha.
Lack of speed of change is not a criticism of quality software. The change of data and how it is handled is important.
Matching the S&P 500 sector percentages is different from building a sector oriented portfolio that will maximize the efficient frontier.
In William Bernstein's first book, "The Intelligent Asset Allocator," he discusses the efficient frontier and available software. I did not find the software all that useful as the portfolios that came out of the calculations were not what any sane investor would put together. Bernstein warns about these rogue portfolios. Collars or constraints need to be placed on the asset classes to produce useful portfolios. I now use a superior program for this purpose.
QPP Analysis Of Portfolioist Portfolio [View article]
Explanation: Seeking Alpha editors pulled the above analysis off my blog (http://bit.ly/rfwO89) and in that process forgot to collect the correlation matrix. That missing data table is important to understanding the explanation of the last two paragraphs.
Does It Make Sense To Include Bonds And Income ETFs In Your Portfolio? [View article]
Lowell
Portfolio Rebalancing: Threshold Limits And Correlation Of Assets Are Determining Factors [View article]
A good example is the Harry Browne "Permanent Portfolio."
Lowell
Portfolio Rebalancing: Threshold Limits And Correlation Of Assets Are Determining Factors [View article]
Good question and I will try to answer as best I can. When faced with a question like this one, I find it useful to set up extremes to help understand what is happening. Here goes. Assume we have an extreme portfolio where 90% of the portfolio is invested in VTI (U.S. Equities) and 10% in BND (broad bond ETF) giving us a stock to bond ratio of 90/10. I ran the analysis moments ago and I see where the correlation between VTI and BND is a very low 8%, or something we might expect. However, in the context of the 90/10 portfolio, VTI has a correlation of 100% whereas BND is 10%. In other words, VTI dominates the portfolio since it controls 90% of the assets.
Now we move to the opposite extreme where only 10% is invested in VTI and 90% is allocated to the bond ETF, BND. While the correlation between VTI and BND remains the same at 8%, that in no way takes into consideration the allocation within the portfolio. In the context of the overloaded exposure to BND, VTI has a 52% correlation and BND an 89% correlation. The construction of the portfolio alters the correlation percentages. Even at 99% allocated to BND and only 1% allocated to VTI, within the portfolio, BND has a 100% correlation and VTI a 13% correlation. These do not need to add to 100%. As the allocation to BND increases, so does the correlation of the overall portfolio. When 90% is allocated to VTI, that one investment essentially controls the movement of the portfolio. When 90% is allocated to BND, it too tends to control the movement of the portfolio, but not quite to the extent as VTI.
While we desire low correlated assets, one only takes advantage of the low correlations if the asset allocation plan devotes a sufficient percentage to the low correlated assets so as to make a difference.
I hope this helps a tad.
Lowell
Does It Make Sense To Include Bonds And Income ETFs In Your Portfolio? [View article]
No, I don't track any of the tickers mentioned in your reply. In no way does that imply your holdings are deficient.
What I do recommend is an accurate method for calculating the performance of the portfolio and all constituent parts. In addition, I'm a strong advocate for benchmarking the portfolio using an appropriate benchmark. I don't think one can easily compare portfolios as each has a different cash flow and launch dates vary.
Lowell
Does It Make Sense To Include Bonds And Income ETFs In Your Portfolio? [View article]
Thank you for the kind words. If interested, check out my blog at this site.
http://bit.ly/rfwO89
I think of it as in educational investment site as I'm trying to help investors avoid the mistakes I made over the years.
Lowell
How To Construct A Customized Benchmark For Your Portfolio [View article]
I both agree and disagree with the first paragraph. In the TLH Spreadsheet I have four benchmarks. 1) VTSMX - as this benchmark will tell me how well I am doing with respect to the total U.S. Equities market. This is similar to you using the Toronto index. 2) VFINX - as this benchmark provides a reference how well the portfolio is performing vs. an investable S&P 500 index. 3) VGTSX or the international market. This index fund provides a reference for how well international investments stack up. This index fund has been lagging the VTSMX by a significant percentage and that helps explain why a portfolio that contains a significant percentage of international securities is bound to lag the U.S. market. 4) The ITA Index or the customized benchmark is the one discussed in the article. This benchmark tells me how well the portfolio is performing compared to my Strategic Asset Allocation plan.
It makes little sense to hold a portfolio of 40% in bonds and then expect the portfolio to outperform the S&P 500 over the long run. Why use the S&P 500 as a benchmark if the portfolio holds a significant percentage in small-cap stocks. While the TLH Spreadsheet provides a comparison with the S&P 500, it provides additional information to the money manager.
As for risk, I use three different measurements. 1) Information Ratio. 2) Sortino Ratio and 3) Retirement Ratio (RR). RR is my own design and a variation of the Sortino Ratio. The SR and RR risk measurements are unique in that they use a semi-variance calculation instead of a mean-variance calculation. The money manager is only penalized with the variation is below the benchmark, not when the volatility is above the benchmark as we want volatility to the upside.
As I mentioned in the article, the ITA Index is not a perfect benchmark, but it goes a long way toward telling the money manager how well they are managing a portfolio based on their Strategic Asset Allocation plan for the portfolio.
Lowell
The Truth About Active Vs. Passive Investing [View article]
I just came across your article when I was searching for Active vs. Passive investing.
Question: Is the S&P 500 the benchmark used in every comparison? If the actively managed mutual funds are making use of small-cap stocks why not use a more appropriate benchmark?
Lowell
Boost Core ETF Portfolio With Low Beta And Piotroski Stocks [View article]
Lowell
Boost Core ETF Portfolio With Low Beta And Piotroski Stocks [View article]
Yes, adding GAI and PFIN is an experiment. I'm using these "Piotroski" stocks with one portfolio. While it is very early in the experiment, here are my IRR results as of moments ago.
CRAI = 16.2% Closed out this position when the stock dropped below 7 on Piotroski's High F-Score.
PFIN = 73.1% Exaggerated value due to short time frame.
GAI = 8958.2% VERY exaggerated due to short time frame.
I check the Piotroski High F-Score screen every week to see if the stock still holds.
Lowell
Swensen Portfolio Supercharged With 2 Piotroski Stocks [View article]
You are not missing anything that has come to my attention. Right now GAI has a High-F score and I will continue to hold it in a portfolio so long as it maintains a rating of 7 or higher. I generally wait to buy when the score is 9, or a perfect Piotroski score.
I picked up shares at $5.15. Just lucky.
Lowell
2 Popular Minimum-Volatility ETF Strategies [View article]
Lowell
QPP Analysis Of Portfolioist Portfolio [View article]
Are Stocks And Bonds Dead? [View article]
Lack of speed of change is not a criticism of quality software. The change of data and how it is handled is important.
Lowell
Are Stocks And Bonds Dead? [View article]
In William Bernstein's first book, "The Intelligent Asset Allocator," he discusses the efficient frontier and available software. I did not find the software all that useful as the portfolios that came out of the calculations were not what any sane investor would put together. Bernstein warns about these rogue portfolios. Collars or constraints need to be placed on the asset classes to produce useful portfolios. I now use a superior program for this purpose.
Lowell
QPP Analysis Of Portfolioist Portfolio [View article]
Lowell