# Henry L. Becker...'s  Instablog

Henry L. Becker, Jr.
Send Message
Henry L. Becker, Jr., CFP® is the editor of Market&Economy blog which takes a panoramic perspectives on capital markets, economics, financial planning & mores. Mr. Becker is a partner and director of research and investment strategy with Lighthouse Wealth Management, an investment advisory firm... More
My company:
Lighthouse Wealth Management
My blog:
@ the Corner of Market \$ Economy
• ##### ETF Sortino Ratios and More 4 comments
Jan 11, 2010 3:46 PM

When you plow the internet for ETF Sortino Ratios you find very little (almost nothing).  If the market crisis we are still living through should have taught advisors anything it should be to take a fresh look at your approach.  I have done just that and started with how I look at risk.

Instead of placing so much emphasis on standard deviation, as most of us have been conditioned to use, I have moved to the Sortino Ratio and downside risk.  In case you are not familiar with the Sortino Ratio it is more or less the Sharpe Ratio except you replace the standard deviation in the formula with downside risk.  The larger the Sortino Ratio the better. So, the Sortino formula looks like this:

Sortino = (Return - Target Return) / Downside Risk

For my calculations the Return is the average return over the last 36 months (where available).  The target monthly return is .48667% (approx. 5.84% annual).  Downside risk is a more complicated calculation but here it is.  First, we identify those returns which are less than the target return each month.  Second,  square those returns that are less than target returns for each month.  Third, calculate the sum of the squared returns from step two.  Fourth, divide that sum by the total number of months.  Last, we find the square root of that number (then multiply it by 100 to get a percentage).

My goal was not to pick apart every ETF available in the US, but to look at countries, global sectors, and some US sectors.  Statistics I calculated for each fund are based on 36 months where available.  Statistics compiled  are:

• Annual Sortino Ratio at 5.84% Target Return
• Correlation to ticker ACWI (iShares All World Equity Index)
• Annual Standard Deviation
• Annual Downside risk at 5.84% target Return
• Mean Monthly Return
• Volatility Skewness (more or less the ratio of months above target to months below target)  Higher the number the better.
• Cumulative Return for Period
Below is an image / link to the PDF results.

Disclosure: IAU EEM EWA LQD TIP EABLX MOO IXN IYM JXI VCR
Back To Henry L. Becker, Jr.'s Instablog HomePage »

Instablogs are blogs which are instantly set up and networked within the Seeking Alpha community. Instablog posts are not selected, edited or screened by Seeking Alpha editors, in contrast to contributors' articles.

• nyguy123456
, contributor

Henry,

Thanks for the article. However, I think these numbers might give unwarranted comfort to someone investing in Gold or Silver. Just take a look at what happened in the late 70s with Gold, and thereafter it returned pretty much nothing for 20 or so years.

In post-modern portfolio theory, downside risk is not measured discretely, it is an integral calculus measure. First, you bootstrap a statistical distribution to the return profile over as long a time period as possible, adjusting skew factor to fit your data better. Secondly, you compute "downside risk" as the squared deviation from the target return, summed over negative infinity (the left end of the distribution) to the target return (zero deviation exists here), and take the square root of the number. The idea here is you're capturing ALL possible returns (hence the use of a probability distribution function), and that would more accurately capture the true downside risk.

I am sure that if you were to calculate this discretely over a long enough time horizon you would get a fairly similar answer, but 3 years is definitely not enough, and I think such a short time window unfairly represents the true risk of each asset class.
15 Feb 2010, 12:22 PM Reply Like
• Henry L. Becker, Jr.
, contributor

Typically, the Sortino ratio is used in place of standard deviation which typically is quoted over 36 months. Hence, my use of the 36 months. You are correct though that longer time will tell a different story. But, the last three years have been interesting upside and downside.

23 Feb 2010, 01:20 PM Reply Like
• Henry,

I too have moved to the Sortino ratio. I was using the Information Ratio, but now have the Sortino ratio built into an Excel spreadsheet I use to monitor a number of portfolios.

An updated list of portfolios tracked using the TLH spreadsheet is available at this site. Check out the performance numbers. Not all are great.

www.lherr.org/blog/

Not only does the spreadsheet contain the SR calculation, but it also has a way to build a customized benchmark for each portfolio. Once I have sufficient data, I will be using this customized benchmark as my target return value, now called Desired Target Return (DRT) by Dr. Sortino.

To learn more about the SR, serious investors will find the book, "The Sortino Framework for Constructing Portfolios" of interest.

Physlab
10 Apr 2010, 03:07 PM Reply Like
• Henry,

Let me correct the blog address as the link above is no longer active. The correct link is as follows.

itawealthmanagement.com

I have the Sortino ratio built into what is called the TLH Spreadsheet. In addition, there is an even tougher hurdle similar to the Sortino and it is called the Retirement Ratio.

Lowell
29 May 2011, 09:18 AM Reply Like