Conditions Of New Bull Market: Coppock Guide
Published May 26th, 2008 in Technical Analysis Tags: bear market, bull market, Coppock Curve, Coppock Guide, Dow Jones, Edwin S. Coppock, false signal, Investech, jim stack, MACD, oscillator, roc, RSI, technical analysts, technical indicator.Continuing with the series, here is the fourth condition of a new bull market as outlined by Jim Stack of InvesTech:
I’ve hesitated to mention this technical indicator since I started writing this blog because it is almost too good. It is one of the few that have an uncanny ability to find the start of almost all major bull markets. So you can understand why I don’t want to run the risk of ruining it by popularizing it any more than it is. And it is not popular at all.
In fact, compared to say the RSI or MACD, the Coppock Guide is an esoteric and rarely mentioned technical indicator. It was created by Edwin S. Coppock some 50 years ago and although it is followed closely by a very small group of technical analysts, its calculation is not complicated at all.
You can keep track of it yourself. Here’s the recipe: you need historical monthly Dow Jones Industrial data. You add the 14 month ROC to an 11 month ROC, then you take a 10 month (simple linear) weighted moving average of the result. That’s it.
If you’re mathematically astute, you’ve already noticed that it is just another oscillator. Here is the chart of the Coppock Guide for the past few years, courtesy of InvesTech:
How is the Coppock Guide interpreted?
The most traditional interpretation is to recognize a buy signal when the Coppock Guide curls up while it is below the zero line.
It can also provide sell signals, although these are less frequent. If the Coppock curve makes a double top formation without first having come down to the zero line (or below it), the market is in for a seriously brutal bear market. You should be able to find one such occurrence in the chart.
So you can see why I think it is almost too good to share. In its history, only 4 false signals have occurred. That’s an 83% accuracy rate.
What is the Coppock Guide saying now?
The good news is that the Coppock Guide is in negative territory projected to fall into negative territory this month. So now any upturn can potentially give us a buy signal. The bad news is that this may happen next week, next month or next year.
The key factor is an upturn. But that can happen from an incredibly low level, like say in 1974 or 1932 (not shown) or it may happen just under the zero line, as in 1994.
Although no indicator can give a full iron clad guarantee, when the Coppock Guide turns up it would totally skew the probabilities towards a new bull market. As always I’m keeping a close watch and now that “the cat is out of the bag”, you can too.
bear market, bull market, Coppock Curve, Coppock Guide, Dow Jones, Edwin S. Coppock, false signal, Investech, jim stack, MACD, oscillator, roc, RSI, technical analysts, technical indicator
Enjoyed this? Don't miss the next one, grab the feed or
15 Responses to “Conditions Of New Bull Market: Coppock Guide”
- 1 Pingback on May 27th, 2008 at 8:06 am
- 2 Pingback on Jun 6th, 2008 at 6:06 pm
Leave a Reply
Search
Creative Commons License
Trader’s Narrative is powered by
Interesting. Though when I do the math the result for the equation is not yet negative for either the S&P or the DJIA.
Jack, to be honest I didn’t double check it but took Stack on his word.
I did the calculation and get a value of 19.68. I use the equation outlined at http://www.safehaven.com/article-1766.htm which I think is the same as the one you outlined in the blog post. I noticed my curve is slightly shifted higher than the curve posted in the blog.
James, thanks for the feedback. I’m going to contact Jim Stack and see if he has an explanation.
Same here, we’re still definitely positive (TradeStation data), at 26.33. Also, with my data, we had no less than 3 false alarms during the tech bear: Apr 01, Jan 02, and Nov 02, before we got the final correct signal. We also had two false signals for each of the 1929 bear market and for the 1937 bear markets.
Thanks for your good research; I enjoy reading your blog!
Thanks b. This is rather strange. There’s got to be a simple explanation for the discrepancies. Perhaps you’re using alternating weights? that is one person starts with 10,9,8, etc and another with 1,2,3, etc? In any case, I’m going to wait to hear back from Stack.
Here is the explanation: since the Coppock Guide uses monthly data we don’t have the data point for May yet, especially not when Jim Stack wrote the latest InvesTech newsletter in early May. So Stack was projecting that the Coppock Guide would fall into negative levels by May. It might and it might not fulfill the projection.
Thanks to everyone who responded and I’m sorry to have inadvertently given a wrong impression. It has been corrected in the post (above).
By the way, I subscribe to Jim Stack’s Investech newsletter. He’s very good with his market analysis, and he’s not afraid to make bold predictions. He’s very cautious, so you have to factor that into your investment decisions, but I’ve found he’s usually right.
J.B.
Actually, I meant to say that he’s not afraid to make bold predictions that go against what most of the other market analysts are saying. He’s not afraid to stick his neck out, and that quality has been extremely valuable to me.
Joe Bob,
you’re right. I also like how he is able to fluidly move between both fundamental and technical analysis. There are very few who use both.
This is the real deal Long term value investor alert.
It is June and this went negative in May. My calculation for May is: -2.3
It has actually had 3 false calls since 1950 — around 07/1957, 07/2001, 05/2002. Other than that it has produced continuous downward movement below zero until it turned up. Then continuous upward movement until above zero.
It does not call the bottom, but shows that a bottom has been put in the DJI and a bull market has been set up.
If you allow 3 months for a bottom signal it has a 100% accuracy rate and has only one failure for a 2 month call.
Also the most significant moves have been after a confirmed upward turn when below zero.
paul, yes, it is probably the key indicator for knowing what kind of market we’re in. I’m puzzled why so many have different numbers and results. anyone care to show their work in a google spreadsheet?
I was trying out the new Excel on a mac — made a error in the data because of the way I imported the csv (had a 10m ROC). Ran it again and got 5.24 for the end of May and .68 for June 4th close. Here is my Excel process.
Download monthly DJI from yahoo (1950 to present month). Delete the final month if you don’t want current day as the last ROC. This may be the data discrepancy.
Upload to Excel and input Formulas and drag them to May 2008:
i. 11m ROC =ROUND(((E13-E2)/E2 * 100),2)
J. 14m ROC =ROUND(((E16-E2)/E2 * 100),2)
k. Add =I16 J16
l. Weighted AVG =ROUND(((K25*10) (K24*9) (K23*8) (K22*7) (K21*6) (K20*5) (K19*4) (K18*3) (K17*2) (K16*1))/55,2)