As promised, here is a link to MarketSci’s review of the ROC indicator. Be sure that MarketSci is on your regular blog-reading rotation, especially if system-trading is your thing. It is absolutely one of my favorite blogs.
A huge thanks goes out to Michael for reviewing this idea.
Woodshedder’s Long-Term Indicator
Comments »Do you feel lucky punk? Do you? That is what I ask myself when I think about dip-buying this market. One of my indicators is suggesting that it might be time to buy the dip for a quick bounce.
Professionals know that mean-reversion works better in volatile markets than it does when markets are trending. The only catch is that sometimes the market doesn’t mean-revert, and you are holding long or short into a death spiral or melt-up. I am fearful that we are again ripe for a death spiral, like we saw in August of this year. However, If I ignore my intuition and focus on the odds, the odds favor a bounce here. Lucky for us, death spirals are uncommon while bounces are common.
This custom indicator simply counts the number of stocks that declined and then ranks the number against the previous numbers of the last 252 days. I call it the Decline Line indicator, or Decliners, for short. There is nothing magical about it, and it is not even really original, but it works. I really like this indicator because it is one of few that do better when the buys are made on the open, rather than at the close when the signal is triggered.
The Rules:
Buy SPY at the open if
Sell the long position at the close if
No commissions or slippage included.
The graph above shows that tonight the indicator again closed above 90. It also shows all of the buy and sell signals for the past year. Technically, the system is long as of Friday’s open, but that doesn’t really matter much to us. We are just looking at probabilities.
Looking at 1.1.2010 to the present:
Looking at all SPY history:
The bottom line is that there is better than a 65% chance that SPY closes higher than tomorrow’s open, sometime over the next week.
Still, despite the probabilities, beware the death spiral. Look at late July and early August for an example.
Equity Curves:
The above graph is from 1.1.2010 to 10.3.11. Note the death spiral drawdown in August.
Note that the indicator has done well during the recent Bear Markets of 2008 and the current one, compared to the Bear Market of 2000.
Comments »Update– As of Thursday’s close, the indicator crossed.
I have it on good faith that on October 1st, one of my favorite system deginger/traders is publishing a review of this indicator. As soon as it gets posted, I’ll link to it here. One thing he has found is that the short signal has not always worked well. It’s success is more of a recent development…Keep that in mind.
2nd Update– As of Friday’s close, the indicator was still crossed. The signal is therefore confirmed. I put on a small short in order to make it real and track the performance.
——————————————————————————————————————-
I introduced this indicator here: Simple, Long-Term Indicator Near to Giving Short Signal.
Just over a month later, the indicator is nearing a cross. See below…
The blue and red lines (252 day ROC and 5 day ROC, respectively) are very near to crossing.
If the cross occurs, I’ll put up a post and may provide further analysis/tweaks to the indicator.
Comments »Definitely not a bullish development, especially considering the EU news is being sold hard. Or is it?
The latest 200 day closing low for IWM is at $64.73. As I write this, IWM is trading at $64.57.
My intuition tells me that buying $RUT on the close when it makes a new 200 day low is a bad idea. However, a quick test shows that my intuition is incorrect.
The first trade was on 3.18.1997.
There were only 8 trades that were held the full 100 days.
These results differ somewhat from my research on buying SPY when it is making new 200 day lows.
Comments »
I was thinking about how to describe my frustration and dis-belief with the Euro-zone financial shenanigans. Lucky for me, Fly had already written just about everything I was thinking.
The action of the past few days reminds me a lot of what happened after September 18, 2008, when Paulson and Bernanke met with members of congress to propose the 700 Billion Bailout (which eventually became TARP). The next day, SPY gapped up over 5.5% but reversed to close up 3.4%. After that day, SPY began falling and almost never stopped until it found a temporary bottom in November.
I’m not saying that is what is going to happen now, but I am very very suspicious of this Germany-Bails out-Europe solution. The mess in Europe seems worse to me than the American credit crisis, and I have a hard time believing that the S&P 500 has found a bottom amidst all the uncertainty. If anything the recent bounce has given professionals one last time to lighten up before new lows are made.
Yes, my gut still says we will see new lows. While the Nasdaq chart is not terrible, the S&P 500 and the Russell 2000 both look awful, like they are just going to fall off a cliff, any day now. There is now strong resistance overhead on both charts, and the falling 50 day average seems to be acting like a lid on price.
Anyway, the last three days closed higher, and gained more than 4%. Lets look at what happened in the past after this setup.
The Rules:
Buy SPY at the Close If:
No commissions or slippage included. All SPY history used.
The Results:
As we would expect, the results show some reversion to the mean over the next week. In fact, during high volatility environments, it works very very well to sell the rips. I fully expect a pullback here, unless the Europeans are able to pull a rabbit out of their silly little bag of magic tricks.
Weird Omen Alert: Anyone notice that the chart above looks almost exactly like the last 2 months of the S&P 500 and Russell 2000?
Comments »On Friday, SPY had traded for 38 days beneath its 200 day moving average. Is this bullish or bearish for the future?
Trading beneath the 200 day moving average (MA200) is considered by some traders to be a reliable indicator of a bear market. The research below confirms that a binary approach to using the MA200 is too simple. Over the past 16 years, the longer SPY trades beneath the MA200, the more likely it is that it is in a bear market. However, long-term mean reversion eventually takes over and the longer it trades beneath the MA200, the more likely it is that the bear market is nearing its end.
I included $SPX in the tests because I have data for it going back 60+ years. As has been a recent theme on this blog, note that the bear markets included in the SPY data (16+ years) have been considerably longer and more volatile than the bear markets over the past 60 years.
The Rules:
Buy SPY or $SPX at the Close If:
No commissions or slippage included. All SPY history used. $SPX data starts in 1960.
The Results:
We want to focus on the dark blue line, which just happens to be the lowest on the graph. That line represents what would have happened historically if Friday’s close was bought (buying after SPY traded for 38 days beneath the MA200). Unfortunately, it seems that SPY is currently in a very bad spot, in terms of how long it has traded beneath the MA200.
Note the green and red lines. They show that if SPY has traded for a short time (10 days) or for a very long time (100 days) beneath the average, results improve. Note how volatile the red line is.
The $SPX data is represented by the light blue and purple lines. The purple line is moderately bullish while the light blue line is bearish to neutral. Over the last 60 years, we see that there have been mild bear markets which have smoothed the averages.
The bottom line is that SPY is currently in a bad spot. It has traded long enough beneath the MA200 to do serious damage but needs to trade longer beneath the average before we start to look for long-term mean reversion to take over.
I’m inclined to believe that the current bear market will look much like the last bear markets of the past 16 years, although it is important to consider that it may turn out to be more mild, and thus we might expect future performance to look more similar to what is suggested by the $SPX data.
Comments »
