Timing attempts to provide market equivalent returns over the long term, with a substantial reduction in variability of returns. The two components of the Timing program are EZ+Macro and Fear/Greed. This system trades rarely and splits its allocations between ETFs tracking the S&P 500, the intermediate-term U.S. Treasuries, and cash.
Information is as of the close on November 23, 2007.
EZ+Macro
EZ Trend is up for the U.S. stock market as approximated by the S&P 500. While the U.S. Ten-Year Treasury price is also bullish by EZ Trend, my latest backtesting hasn’t indicated that a position in bonds is warranted in this EZ+Macro configuration.
Fear/Greed
The Fear/Greed model signaled a buy for the U.S. stock market in early November. It would signal a sell only if $VIX relative to actual volatility fell to a historically low level. This is a tough model to follow, as it demands a buy and hold when fear is high and most people would like to sell.
Model Allocation
S&P 500 SPDRs (SPY) – 100%
iShares 7-10 Year Treasury Bond Fund (IEF) – 0%
Cash – 0%
Tracking
There are no changes to the allocation from last week’s message.
Charts and Commentary
My charts are 700px wide and this site is best viewed at 1280×1024. If the chart is truncated in your browser, click to view it in full size.
(1) EZ Trend components. This is the lowest the spread has been in the past 52 weeks, but the signal is a cross of the averages, not a spread distance.
(2) Fear/Greed buy signal. As long as options estimates of S&P 500 volatility are spiking in relation to actual volatility, this system will be on a “buy” signal. Monday’s market drop may have been distressing to many long traders, but that very distress was a bullish signal by this method.
(3) This week, the price action crossed below the congestion period marked during the July to September correction, and closed above it on Friday afternoon. The index stayed well above the intraday lows of August 16, but just about nicked the closing values of August 15th and 16th. This might be considered a successful retest of the lows by some technicians.
(4) Volume spike, with a long lower wick on the candle chart, on July 26. Lower prices were tested and the index bounced back to a large degree, then fell more on the following day with volume that was slightly lower, but still significantly above the average volume at that time. Most of the price action in the remainder of the correction stayed near that second day’s price range.
(5) Looks familiar to me. No guarantee that it will play out the same, but I’m not thinking there’s much downside here.
(6) 2-day RSI is neither oversold nor overbought. Any short-term signal remains days away, depending on the methodology used.
Note the breadth in the following charts.
The weekly candles for breadth measurements on the 50- and 200-day simple moving averages appear to have printed hammer patterns at support. Bullish percent is the lowest it’s been in the last few years, barring the week of August 16th. The Bullish Percent Index (BPI) is a popular market breadth indicator that is calculated by dividing the number of stocks in a given group (an exchange, an industry, etc.) that are currently trading with Point and Figure buy signals, by the total number of stocks in that group. I remain likely to consider very low breadth as a contrary buy signal. There’s no guarantee that the members of the index won’t become more oversold as a group, but I view it as unlikely to happen here.
Regression models show 90/100 for 20-day “potential” and 70/100 for 20-day “safety.” Twenty trading days is approximately a month. These are bullish values. I don’t include the regression model in the timing system because I specifically want the system to be “lazy” i.e. not trading very often.
