Signal Type	Trade Date	Return since issued
		World	U.S.
			Nasdaq 100 (QQQQ) Russell 2000 (IWM) S&P 500 (SPY)

Market Update

It has been a week marked by low trading volume, in which the main indexes posted modest losses overall. Stocks started by skidding the first two days of the week. Monday's losses were fueled by concerns over the future of Freddie Mac and Fannie Mae, as both companies saw their stock plunge more than 20% on talks that a bailout from the U.S. treasury might be required to save them. The S&P 500 lost 1.5% on the day. Stocks lost more ground Tuesday as inflation fears resurfaced following the release of the producer price index for July: the core PPI, which excludes volatile food and energy costs, rose 0.7% last month, more than economists expected. Rumors that Lehman Bros might announce a huge write-down this quarter and a disappointing earnings report and cautious outlook by Home Depot also contributed to the negative tone. Helped by bullish earnings news from Dow component Hewlett-Packard, stocks were able to rebound Wednesday, with the S&P 500 posting a 0.6% daily gain. The main indexes proved their resiliency Thursday by recovering from an early drop caused by higher oil prices to close in the green or with only small losses. Showing renewed strength, stocks finished the week by closing solidly higher on Friday, with the Nasdaq Composite gaining 1.4% during the session. Investors were encouraged by lower oil prices and comments from Fed chairman Ben Bernanke, who stated that inflationary pressures should moderate this year.

For the week, the S&P 500 (SPY) , Nasdaq 100 (QQQQ) and Russell 2000 (IWM) respectively lost 0.40%, 1.37% and 1.89%. The Nasdaq 100 and the Russell 2000 remain located above both their 50-day and 200-day exponential moving averages (EMAs), while the S&P 500 is still situated above its 50-day EMA but remains under its 200-day EMA.

For its part, our World portfolio outperformed its U.S. counterparts this week with a 0.24% loss. The portfolio consists of the 5 top-ranked world ETFs as of August 15, which marked the beginning of the current 4-week holding period. Please go to the "Our Service" page for all the details.

Our current Buy signal remains in effect.

Trend Timing School

Annualized versus cumulative and yearly returns

We regularly receive questions about the different kind of return definitions. Some of our readers get confused with all the terminology and, even more, by the respective calculations. The good news is that our results have been checked and re-checked thousands of times and no one has ever found a problem with them. So, unless you really feel an urge to brush-up on your math skills, you can leave the heavy lifting to us. You do not have to be a math major to be a good Trend Timer!

For the curious minded, here are the definitions of terms frequently encountered with investment returns.

Annualized return
This is the one number which, if used as the return each and every year during a multi-year time period, would have resulted in the cumulative return over that period, as if it had grown at a steady rate.

Annualized return formula:
   annualized return = ((1 + cumulative return) ** (1 / years)) - 1
where * * denotes exponentiation or power of

Don't worry; it gets simpler with real numbers.
For example, we invest $10,000 for 2 years.
The first year we have a great return of 50%, the $10,000 grows to $15,000.
The second year is not nearly as good and our investment loses 10% to end up at $13,500. The multiplier is 1.35 which means that the cumulative return is 35%.
To get to the annualized return we need to find the one consistent return number which when used during the 2 years would have grown the $10,000 to $13,500. Using the formula above, we take 1.35 (1 plus the cumulative return) to the power of 1/2 (which happens to be the same as the square root) which is equal to 1.162, minus 1 which gives us an annualized return of 0.162 or 16.2%.
To verify, $10,000 times 1.162 is $11,620 times 1.162 is $13,500. Bingo!

So why not simply use the average yearly return instead, you may ask.
Because it does not get us to the correct cumulative return.
Using our previous example and the average yearly return formula below, the average yearly return is 0.50 plus -0.10 equals 0.40, divided by 2 equals 0.20 or 20%. But $10,000 times 1.20 is $12,000, times 1.20 is $14.400, which is obviously the wrong answer, and this is why they invented the nifty annualized return. Just one more reason why we feel yearly returns are misleading.

Average yearly return
This is the simple arithmetic average of the yearly returns over several years. We do not favor or use this metric for the same reasons we do not favor the yearly returns.

Average yearly return formula:
   average yearly return = (yearly return 1 + yearly return 2) / 2

Compounded return
This is identical to the cumulative return explained below

Compounded Annual Growth Rate (CAGR)
This is identical to the annualized return explained above

Cumulative return
We like this metric as it best represents what an initial investment will grow to over a period of years, because it factors in the compounding effect of reinvested gains. The returns of individual trades during the entire period are compounded (not added) using the following formula.

Cumulative return formula:
   cumulative return = ((1 + trade 1 return) * (1 + trade 2 return) * . . . * (1 + trade n return)) - 1

Yearly return
The investment return in a given calendar year. Unlike annualized return, yearly return represents the actual performance for the year. This is very similar to the Annual Percentage Rate (APR) you hear about when you take out a loan. The primary difference is that you pay the APR interest but you receive the yearly return. We do not favor yearly returns because they are misleading and totally ignore the compounding effect so important to long term Trend Timers. Because it is a widely used financial metric we do publish our historical yearly returns. They can be found in the Yearly Returns section of the Results page

Yearly return formula:
   yearly return = (end price / begin price) - 1

Congratulations to the few valiant souls that have made it this far without dozing off!

FAQ of the Week

Question: What will be the market's next move?

Many investors, new subscribers in particular, express frustration at not knowing what or when the market's next move is going to be. The natural investor predisposition is to want to analyze, think, guess, and outsmart the system. It can sometimes be tempting to liquidate our positions and take some profits? It is certainly for each individual to decide, but this is not how Trend Timing is designed to work.

The black box nature of our Model understandably exacerbates the feeling of blindness when in fact it distinctively illuminates the path ahead. The truth is that we do not know how long the trend is going to last or how strong it is going to be. No one can. The Model tells us what the predominant market trend is at the current time, and as long as this trend is in place the odds favor a continuation and not a reversal. No market goes straight up or straight down for any length of time. There can be pullbacks and corrections in the trend, but following an urge to bail out at some stage could leave you stranded on the wrong side of the trend, usually not a very good place to be.

The current market choppiness does not make things easier, but since we cannot predict the future, let's just sit back, relax, and follow the trend.

Warm wishes and until next week.

The TimingCube Staff