Single Country International Strategy
Building on the Sector Strategy previously published we will now develop a similar strategy for trading international ETFs. If you have not read about the previous strategy I highly recommend following the above link and reading that page before this. Why? Because many of the same concepts will be used here, but without the full descriptions. This will give us two strategies aimed at differing universes of funds that stretch around the globe.
There are several ways to achieve international exposure using ETFs. There are worldwide funds, regional funds, and country specific funds. Here we are going to use the iShares family of MSCI single country funds. This gives us more independence between funds and we don't have exchanges between highly correlated funds, like from Latin America and into the Brazil, where any significant move in one will most likely be mirrored in the other. At this time there are 21 such funds. Most began trading in 1996 with a few others in 2000 and the Singapore fund in 2003. SPY was added to include the U.S. market in our study and two Russell value funds (IWD and IWN) were added to provide some buffer during bear markets. I would have included some bond funds in this mix but none of the bond related ETFs have been trading long enough to be of benefit during this backtest.
Like the first strategy, this is a trend following system using the 6-month trailing return as the signal. Since this is a worldwide strategy with plenty of opportunity for diversification this model is designed to hold three postions at a time, where our prior Sector Strategy only holds two. And since we are holding more positions we will increase our sell point from rank six to rank 7. That is, we will buy from ranks 1 thru 3, but only sell a position if it falls below rank #7 when we sort by 6-month return.
The procedure will be like before. On the first trading day of the month we will calculate the 6-month total return for each of the above ETFs. This data can be optained from several sources including this web site by using the Current Rankings link near the top of this page. We will sort the ETFs from highest to lowest return value. The first month we will purchase the first three ETFs on the list. After that we will sell an ETF if it is not in the top 7 on the list and replace it with the top fund we do not already hold.
Table 2 shows all the trades in our backtest along with the dates and returns for those trades. The test began on 1/2/2001 and ended on 11/30/2005. You can see that 19 of the 26 trades were positive with most trades outperforming the SPY during the same time period. We were hurt by the start date of this backtest. Had we been invested prior to 2001 we would have taken the Switzerland fund in September of 2000 before the peak and would never have taken the position in Italy. But, we had to start some time and 2001 was the logical point given that the Russell funds used in our test did not begin trading until early 2000.
The system produced a 17.4% CAGR over this backtest period while the market returned an annualized 0.9%. As always, there is no guarantee that future performance will be like past performance, but we have a good starting point. More stats are in Table 3.
Month by month performance data for the model and the SPY are in Tables 4 and 5. As you can see from this data our simple model has outperformed the broad market in 37 of 59 months and for every year of our test period.
The graph below shows this model more than doubling in value during a time when the market was basically flat. It also shows that during a bear market that even the international markets can be a difficult place to invest. But as the markets righted themselves, this model pulled ahead and has performed well ever since.
This article shows how a basic trend following system can outperform the market. The six-month returns are calculated by measuring the total return (including dividends) over the past 126 market days. This simple calculation can be performed with a calculator or the return data can be obtained from this web site. If you have any questions or comments feel free to post them using the form below, or email me using the link at the bottom of the page.
Disclaimers: Note that these results should be considered hypothetical, that the closing prices used may not have been achievable in real trading, and that no transaction fees are included. Also recognize that past performance does not necessarily predict future performance and therefore you should perform your own due diligence before following this or any other investment strategy.
Current Comments195 comments so far (post your own)
© 2005 H.M Todd