All donations go towards web site maintenance for all of Technical Watch,
keep it free of charge, and may be tax deductable as an investment expense.


PayPal Verified
Join our market chat sessions every Tuesday and Thursday at 4:00 pm Pacific time!
More information on subscriber services can be found at
http://www.technicalwatch.com/subs.htm

Sign up Calendar Latest Topics
 
 
 


Reply
  Author   Comment  
mortiz

TW Patron++
Registered:
Posts: 1,054
Reply with quote  #1 

The Hindenburg Omen (HO) has been speaking again since early April, and is getting a lot of attention among bulls and bears alike.

 

Robert McHugh has written a couple of articles on the HO, with background on the omen at this link:

 

http://www.safehaven.com/article-4937.htm

 

with an HO update at this link:

 

http://www.safehaven.com/article-5019.htm

 

Below is a summary of the HO criteria:

 

1)     The daily number of NYSE New 52 Week Highs and the Daily number of New 52 Week Lows must both be high enough as to have the lesser of the two be greater than 2.2 percent of total NYSE issues traded that day.

2)     NYSE 10 Week Moving Average is Rising

3)     The McClellan Oscillator is negative on that same day

4)     Added Filter: New 52 Week NYSE Highs cannot be more than twice New 52 Week Lows, however it is okay for New 52 Week Lows to be more than double New 52 Week Highs

5)     Added Filter: A confirmed Hindenburg Omen, in other words for it to be "official," there must be more than one signal within a 36 day period.

 

For For those who read the background and consequences of the HO appearing in the market, you know the potential devastation on prices.  However, like any indicator, the HO is not perfect.  According to my data, the HO criteria has been met seven times since April 7, 2006 using the NYSE composite new highs-new lows numbers (all issues), with the high-low criteria met by the common only high-lows on three occasions in April.  Many of the new lows posted with the composite data is due to preferred stocks and other interest rate sensitive derivatives traded on the NYSE.

 

       Out of curiosity, I took a look at the state of liquidity, as measured by the NYSE common stock AD line, during the times when the HO criteria were met to determine if there was any correlation between the HO's impact upon prices and liquidity.

        

       First, is a chart comparing the NYSE composite index (NYA) against the common stock only cumulative AD line.  Circled in red on the blue AD line are the locations where the HO signals were confirmed per Mr. McHugh's study. Note this version of the common only AD line includes all of the legacy method common only data as reported by the NYSE from the early 1980s until February 2005. 

 

      In early 2005, the NYSE changed it methodology in deriving the common only AD data to a more pure approach that included only the commons.  Prior to 2005, the "common data" included many issues, from 600 to 700 at times, including CEFs, ETFs, and some structured products.  therefore, the common data since early 2005 has been "back-adjusted" to mimic the legacy method for consistency for a historic look at the data.   

 

  

 

       To more easily identify the HO events, a "zoom in" on a few different time frames will provide more clarity in comparing price and liquidity (AD line) during the HO events.

 

       First is the 1983 to 1990 window.  The first HO from Mr. McHugh's study was confirmed in mid-July 1986.  The McHugh study used the Dow as the price proxy, this study uses the NYA since its price behavior is more representative of the whole common stock market.  The 1986 HO resulted in an 8.5% NYA decline.  Note the common only AD line had achieved a new three year high three months prior to the HO appearing.

 

      The famous 1987 crash lost 29% in NYA price following the HO criteria being met. Leading into the crash, the common AD line had fallen below its 1% trend (199 day EMA) following nearly 18 months of chopping within a range.  In addition, a new sheriff was in town named Alan Greenspan who immediately marked his new territory with a large increase the Fed target rate.

 

       The 1989 HO event was followed by another 9.8% decline after the AD line failed to exceed its 1987 high and had fallen below its 1% trend.  Sorry about the circled locations on the AD line getting corrupted in the images, but they are close enough to the exact zone to provide a feel of the AD line configuration at the time of interest.

 

  

    

       Next is the 1990 through 1995 time frame.  The first HO in mid-1990 resulted in a 17% price decline.  At that time, the AD line was in a pronounced downtrend, indicating liquidity was not strong.

 

       There were then two failed omens, in late 1991 and late 1993, where the NYA price declines were in the 1% and change variety.  In January, 1994, the AD line was in the process of achieving a new 41 month high when the HO surfaced, followed by a 7% NYA price decline, not exactly a crash, but enough carnage where most long positions would have felt the hit.

 

      In September 1994 the HO appearance resulted in a 6% NYA price decline while the AD line was in an obvious down trend.  Then another failed HO in October 1995 with very little price impact.  Note the AD line was initiating a new long term uptrend during the failed HO, illustrating liquidity was strengthening again.

 

 

      The 1996 to 2000 time frame had its first HO surface in June 1996 with a 5.6 % price decline following a new multi-year high in the AD line.  Again, not a crash, but moderate price damage.  In December 1997, a HO appeared followed by a 4% NYA price decline.  The AD line was consolidating before the run into its 1998 high, suggesting liquidity was still ample enough to mute any price declines.

 

      In July 1998, another HO was posted resulting in a serious 19% price decline.  The AD line topped out 3.5 months prior to the price break, falling below its 1% trend, rallying back to the EMA, then collapsing with price after failing to exceed its 1% trend.

 

     Beginning in mid-1999, a failed omen despite weakening liquidity, then three omens in 2000 during the volatile topping process.  The common AD line was dragging across its lower range throughout 2000, showing very little life and a lot of weakness.

 

 

      The final chart illustrates price and AD line action during omen signals from 2001 through the present.  The first HO signal occurred in March 2001, with a modest 6.5% price decline with the AD line remaining anemic.

 

      In June 2001 and June 2002, two omens surfaced during the heart of the great bear market, and the omen's worst case ramifications were hitting on all cylinders with 18.8% and 20.6% price declines following HO confirmations,  The price carnage was actually worse than that if measured from the actual price tops, but all price declines are measured from the first HO confirmation.

 

      Following the end of the bear, two omens have played out, one signal in April 2004 followed by a 5.4% price decline, and another signal in September 2005 resulting in a 4.4% price decline. In the latter event, the common AD line topped out before prices, but the AD line was only catching its breath, liquidity was still strong, and price damage was muted.

 

 

       The current Hindenburg Omen is still in the process of digesting its confirmation signals, and thus far there has been no price damage in the NYA, actually posting new highs last week.  However, the common AD line nor the unweighted NYA is confirming the new weighted NYA price highs, thus any trouble is still ahead. 

 

      With the current configuration of the common AD line, there is no evidence yet the strong liquidity enjoyed by the bulls over the past three years is weakening significantly.  Therefore, the benefit of the doubt must be given to the longer term bullish case, and any price decline in the coming weeks will likely be of the milder sort, barring any geo-political or natural disasters. One could expect price declines to be less than 10%, and likely in the 4% to 6% range at worst, given the current status of the common stock AD line.

 

      One very important data point that must be kept in mind, is the common only AD line is near its all-time highs per the available data since 1980.  The composite NY ratio-adjusted AD line is at its highest level since the all-time highs posted in 1959, thus the liquidity action as measured by the AD line variants, is at historically strong levels, and any price declines will likely be contained until these AD lines begin diverging signficantly... which has yet to occur.

 

     FWIW

 

     Randy

0
geosing

TW Member
Registered:
Posts: 356
Reply with quote  #2 

Excellent study, as always, Randy. Question re divergence between RA-A/D and regular A/D line ... will this be: A/D keeps rising, but RA-A/D does not rise as much or actually starts declining? Is that what one looks for to judge whether the current liquidity flood is abating and a serious correction is in the offing?

 

Mild correction of 4% to 5% is still a substantially profitable one, if one can position funds correctly. e.g. Rydex dynamic funds will yield 8% to 10%, if one goes for less of a leverage than offered by futures. May be option players can do better.

 

Geo

PS: Did you come to any conclusion on your study of weekly A/D as compared to the daily, which da cheif made reference to some time ago?

0
mortiz

TW Patron++
Registered:
Posts: 1,054
Reply with quote  #3 

Geo,

 

I apologize for not clearly distinguishing between the "raw" and "ratio-adjusted" cumulative AD lines.  The raw AD line is just that, 1000 advances, 500 declines, we have +500 net advances.  Due to the the ever increasing number of issues trading on the NYSE, historical comparisons (greater than 5 or 6 years) of the AD line should include the ratio adjusted method, which is the approach used for ratio adjusted McClellan indicators:

 

((Adv - Dec)/(Adv + Dec))*1000

 

Using the raw AD data, the NYSE AD line has easily left the previous 1959 all-time high in the dust some time ago.  In 1959, there were usually 1100 issues traded on the NYSE while today, 3300 to 3400 issues trade on the NYSE on any given day, thus the need for adjusting via the ratio method.

 

Using the ratio adjusted approach, 1959's high is still intact, but that high is only the all-time record on an adjusted basis which is interesting trivia, but how important it is in the current climate? Not too important.  What does matter is the health of the AD line from a liquidity perspective, and its position with respect to its 5%, 10%, and 1% trends.  As we are aware, the difference between the AD line's 10% and 5% trends is the value of the AD McSum, so there is a lot of information in an AD line chart that includes its 5% and 10% trends.

 

When the AD line begins diverging for an extended period of time with price, is where the warning flags rise suggesting liquidity in the markets is waning, and the clock is ticking for a more serious price decline.  Recently, we saw a few divergence in the common AD line with prices which was remedied this past Friday, with a decisive breakout in the common AD line.  We can now reset the "divergence clock".

 

I agree there can be plenty of money made on the short side, with or without leverage in a 5% market price decline.  Many longer term investors are willing to sit out a muted decline (less than 10%), thus the motivation behind the Hindenburg Omen study with respect to liquidity, as measured by the AD line.  For short term traders and intermediate term traders, which I consider the class I fall into, a 5% decline is definitely worth trying to get a piece of in my speculative accounts.  However, for core accounts where investment options are limited, sitting out a 5% decline with an otherwise favorable long term outlook is acceptable to many.

 

As far as the weekly common AD data, I am slowly but surely working on it.  It not a trivial project since I am working with daily files where new issues come aboard, others fall off, not every issue trades everyday, etc. Due to my penchant for accuracy, it is a painstaking project, which I am considering writing a software application to finish it off, but even that will require a non-trivial investment of time.

 

The common data, weekly or daily, tells a different story than the composite AD data, during the bear market, and even right now.

 

Thanks for your feedback Geo, I have to get off this board and get to work!!

 

Randy

 

 

0
Previous Topic | Next Topic
Print
Reply

Quick Navigation:

Easily create a Forum Website with Website Toolbox.






Copyright 2000-2019 Technical Watch