Signal Type	Trade Date	Index	Return since issued
		Nasdaq 100
		Russell 2000
		S&P 500

Market Update

As had been widely anticipated, the Fed decided to leave interest rates unchanged during its Wednesday meeting. After having struggled Tuesday following a disappointing sales forecast from Yahoo!, stocks rallied on the Fed news to send the S&P 500 within a fraction of its yearly high. The gains did not last, as the major averages moved lower the last 2 days of the week after the release of the Philadelphia Fed index reignited fears of an economic slowdown. The Philly Fed index came in at -0.4, well below views of a 14.8 mark. The negative reading indicates that regional manufacturing declined in September. Investors were taken by surprise and reacted by selling shares.

The Nasdaq 100 and S&P 500 respectively lost 0.62% and 0.37% on the week. The Russell 2000 fared worse as it closed 1.48% lower. All three indexes still rest above both their respective 50-day and 200-day exponential moving averages (EMAs). Our Buy signal remains in effect.

Trend Timing School

World Index Ranking statistics

First of all we would like to thank all of you for the extensive feedback you have given us on the World Index Ranking service announcement, and we sincerely appreciate both the positive and the critical kind. We may not always be perfect, or even right, but one thing about the TimingCube team is that we listen to our subscribers' feedback and, just in case you did not receive the e-mail on this important process change, we rapidly modified our sign-up approach from having to opt out to having to opt in.
Please note that you do not have to opt in to gain access to the service during the tryout period, and that you can see the current World Index Ranking right now on the "Current Signal" page. Note that you can opt in and out of the World Index Ranking service at will until October 15, 2006.

Other frequently requested items were more performance statistics, which we are happy to deliver and review below, enhancements to the World Index Ranking table on the "Current Signal" page to make it easier to understand and the addition of a column listing the ETF of choice for each index.

Implementation related question were numerous, in particular "how important are the specific variables of the strategy, such as 4-week periods and 5 index portfolios?" Our backtesting showed the best results for the approach we recommend, but also that variations are acceptable. For example, using only the three highest ranked indexes on the list instead of five did not change the results significantly. While in theory placing all your bets on the top ranked country would seem the best approach we do not recommend any fewer than three indexes for diversification reasons. Rebalancing more often than 4 weeks did not seem to increase performance.

Another frequently suggested strategy for the Sell part of the cycle is to short the bottom five countries on the rankings, following the logic that the weakest countries should perform the worst during down turns. This reasoning is flawed in a couple of ways, starting with the fact that most of these countries/ETFs cannot be shorted. Since the TimingCube Model does not seek to find the geography which is likely to decline the most during downturns, we simply recommend shorting the QQQQ (but it is easier and cheaper to simply buy the short ETF ticker symbol PSQ ).

We also had many inquiries regarding which indexes we selected to include or leave out of the World Index Ranking, as well as which investment vehicles are available to implement the strategies. This is a very important and extensive topic we reserve for future discussion, but in the mean time we address the mutual funds question in this week's FAQ (see below).

Finally we come to the performance statistics which everyone wants to see in order to better evaluate and assess the World Index Ranking service. We have included below additional performance data from our backtesting in the form of Statistics and Yearly Results. Note that we have also added these statistics in the "Performance with World Index Ranking" section of the "Results" page in the form of the "Statistics" and "Yearly Returns" buttons. Since the World Index Ranking Model is designed to be implemented with rebalancing periods of 4 weeks, most of the statistics, such as "Average period return", "Winning periods ratio", and "Average periodic turnover" are measured in 4-week periods (which explains why we cannot easily compare these statistics with the "TimingCube Signal only" approach which goes strictly by signals, not periods).

Over the testing period (from 01/02/2001 to 9/15/2006) applying both World Index Ranking and the TimingCube Signal with the Long and Short strategy greatly improves the return (1086.99%) over the straight TimingCube Signals Long and Short strategy (830.28%). Of course, in our comparison the difference is that one invests in the rotating top 5 world indexes during Buys and short the Nasdaq 100 during Sell signals with the former, versus Nasdaq 100 both long and short for the latter. Being in the strongest markets during broad market up moves is clearly an advantage. At the same time, the "Maximum equity drawdown" came down as well to a manageable (by most) -14.26%.

Another useful statistic is the "Average periodic turnover" which in plain English means how many positions change each period. The Buy and Rebalance strategy shows a turnover of 19.7% which signifies that on average, only about 1 index changes in the top 5 every 4 weeks. This goes a long way to satisfy those of us who were worried that the monthly rebalancing would involve a lot of trading and high commission expenses. The other two World Index Ranking strategies show turnovers of 34.2% which include both the 4-week rebalancing and the TimingCube Signals.

The Yearly Returns clearly show that the winning strategy is to combine the traditional TimingCube Signal with the geographic targeting of the World Index Ranking models, as it delivers the best return three out of five years, and comes a close second the other two. Also telling is that our unofficial 2006 year-to-date figures show the two World Index Ranking strategies, Long Only and Long and Short, deliver 19.35% and 32.86% respectively, compared to 14.99% for the straight TimingCube Signal, with the current Model of course.

Statistics since 2001 (1/2/2001) as of 9/15/2006
	World Index Ranking and TimingCube		World Index Ranking  only	TimingCube Signals  only
	Long Only	Long and Short	Buy and Rebalance	Long and Short
Cumulative return	321.76%	1086.99%	126.76%	830.28%
Annualized return	28.64%	54.17%	15.40%	47.74%
Average 4-week period returns	1.97%	3.68%	1.23%	NA
Winning periods ratio	85.71%	77.63%	68.42%	NA
Average winning period return	3.27%	5.98%	3.65%	NA
Average losing period return	-1.33%	-4.81%	-4.23%	NA
Largest winning period return	9.79%	38.10%	9.79%	NA
Largest losing period return	-4.48%	-13.02%	-18.08%	NA
Maximum equity drawdown	-4.48%	-14.26%	-27.50%	-22.80%
Average periodic turnover	34.2%	34.2%	19.7%	NA

Note: The TimingCube Signals only results shown are for the Nasdaq 100 and a Long and Short strategy.

	World Index Ranking and TimingCube		World Index Ranking only	TimingCube Signals only
Year	Long Only	Long and Short	Buy and Rebalance	Long and Short
2006 as of 9/15/2006	19.35%	32.86%	9.40%	14.99%
2005	38.40%	50.62%	32.93%	22.76%
2004	29.47%	34.59%	23.03%	23.47%
2003	59.18%	61.59%	45.98%	50.07%
2002	2.62%	34.84%	-12.05%	57.28%
2001	20.73%	102.26%	-1.29%	110.69%

Note: The World Index Ranking system was introduced on September 15, 2006; earlier returns are backtested hypothetical results. The tables above provide the historical backtested results for our sample 5-index portfolio. The sample portfolio was started on January 2, 2001 and is rebalanced every 4 weeks. The Long Only strategy invests in the top 5 indexes during Buy signals and goes to cash during Sells. Long and Short returns are for the top 5 indexes during Buy signals and short Nasdaq 100 during Sell signals. The Buy and Rebalance strategy ignores the Buy/Sell signals altogether and stays fully invested in the top 5 indexes.

FAQ of the Week

Question: Can I use mutual funds with the World Index Ranking system?

You might try, but we would not recommend it.

While there are hundreds of international mutual funds on the market today, the vast majority of them are actively managed, meaning that they do not track a stock market index. Further, many of the funds are rather vaguely targeted to regions such as "new Asia" or other groupings like "emerging markets", and in the end there are only very few country funds which could be associated with the World Index Ranking. Another important drawback of mutual funds are the short term redemption fees which are becoming rather pervasive in the industry and which for all practical purposes eliminates them as viable investment vehicles for any sort of investment strategy involving timing or rebalancing. Some mutual fund companies such as ProFunds and Direxion are making strides in enabling trading strategies with international bull/bear mutual funds, but there are only a few selections available and they do not match the country indexes we track, except for Japan (see "Investing with long/short index mutual funds")

We feel that index ETFs are the choice investment vehicle for this type of investing because, in addition to the broad selection of international funds, they offer the convenience of trading like stocks and low cost.

Warm wishes and until next week.

The TimingCube Staff