It seems you have JavaScript disabled.

Ummm.. Yeah... I'm going to have to ask you to turn Javascript back on... Yeah... Thanks.

Golden Cross: Bullish Technical Formation




Last week there was a lot of chatter about a technical formation called a 'golden cross' which is considered to have bullish implications. This is when a short term moving average (usually a simple 50 day MA) crosses from below to rise higher than the long term moving average (usually a simple 200 MA). Because moving average tend to move in lethargic arcs, these types of formations are easy to foresee.

In keeping with everyone's watchful expectation, the S&P 500's 50 day moving average closed at 900.54 on June 24th 2009, rising slightly higher than the 200 moving average (897.19).

Since we've compared the current market to the nascent 2003 bull market in many different ways: breadth, wedge formation, flag formation, Weinstein analysis, etc. It is only natural then to take a look at the golden cross that presaged the bull market in 2003:

golden cross S&P500 May 2003

In the charts, the blue line is the 50 day moving average and the red line is the 200 day moving average. Marked by the green arrow, the medium term moving average crossed higher than the longer term moving average in May 15th, 2003.

But does the golden cross really deserve its bullish moniker? Obviously we can't base any conclusions on one single observation in 2003.

Vincent Delisle of Scotia Capital looked at 14 previous S&P 500 bull markets (lasting on average 49 months and rising 149%). From these only about 17% of the gains materialized before a golden cross signal was given. After 12 months of a signal the average gain was 23%, implying that a golden cross doesn't arrive too late to provide forward returns. Delisle adds that a golden cross appears to have more validity when it occurs with a rising 200 day moving average - something we had in 2003 but do not have now.

golden cross S&P500 June 2009

By the way, a "death cross" is the opposite and can be seen on the above chart marked by a red down arrow.

According to Jason Goepfert of SentimenTrader, any edge offered by golden crosses is minimal. Identifying the same distinction as suggested by Delisle, he looked at only instances where the 200 day moving average is declining.

Goepfert concludes:

...the returns going forward, up to six months later, were little better than random and not statistically significant. In fact, in the shorter-term they were a little worse than random. Only when we look out a year do we see some out-performance.

But he does agree with Delisle that most 'unsuccessful' golden cross signals coincide with the early 1940's and that more recent examples have had much more success. The S&P 500 was positive 11 out of 13 times since 1942 with an average annual return of 18%.

Finally, a reader was kind enough to forward a recent research report from Merrill Lynch on golden crosses. I've added it to the Free Trading Resource Section and you can download it from the Articles & Reports folder.

In the Merrill Lynch report prepared by Mary Ann Bartels, it continues to distinguish between golden crosses that happen with a downward long term moving average and those when the long term moving average is rising:

Of the 42 Golden Cross signals triggered since 1928, 20 have occurred with the 200-day moving average in a declining trend or lower than it was 30 trading sessions ago. These signals on average have generated 12-month returns of 13.3%.

The remaining 22 signals occurred when the 200-day moving average was rising or higher than it was 30 trading sessions ago. The returns for these signals were much lower and on average generated 12-month returns of 5.7%.

This is bullish for today's market since the long term moving average of the S&P 500 is still falling. The report is full of insight backed by stats so I highly recommend you download it and take a look. Bartels also adds a new overlay by looking at golden crosses that happen during a recession (as defined by NBER). Signals that meet the condition of a declining 200 MA and a recession suddenly produce an average 12 months return of 23.3%.

Not surprisingly, her conclusion is that "the equity market remains in a base-building process that should lead to higher returns."

Of course, that doesn't mean that the market automatically heads higher and higher from here. Base building can be soul crushing. Ask any trader that lived through the 1970's - no wonder everyone started wearing platform shoes ;)

Technorati , , , , , , , ,

Enjoyed this? Don't miss the next one, grab the feed  or 

                               subscribe through email:  

16 Responses to “Golden Cross: Bullish Technical Formation”  

  1. 1 Steffen

    Greetings from Germany,

    here a link to an essay posted in a big german stock trading forum. The text is german unfortunately, but the chart should be relevant to the topic of this thread

    The author of this article compares the Nasdaq from 1999 on with the Dow Jones from 1928 on. He argues that in both cases a huge speculative bubble bursted (in 1929 and 2000), and in both cases we are in a 'post-bubble-burst' phase of the market, which unfolds for more than a decade.

    The similarity of the Nasdaq and the Dow Jones seems to be significant. When the markets now follow a similiar pattern as in the 30s, we are in a phase comparable to the years 1937-38. There is also a fundamental similarity: Roosevelt cut massively government spending in 1937 (the crash of 1937 is called the 'Roosevelt depression'). From perspective of money supply, this is comparable to 2007, where suddenly a lot of liquidity was withdrawn from the markets.

    That means, we should expect around 4-5 years more of a dull sideways market, and at around 2012 a new bull market comparable to that of the 50s could begin. This also means that we are now entering a phase where the golden cross signal mostly fails, as it was in the years around 1940.

    Some more remarks on the big picture:

    For long-term investors, it might be a good idea to wait some more years. Any long-term sustainable bull market should be supported be a fundamental reason. I don't still see anything worth, but there might be something at the horizon: The long-term rebuilding of the energy supply. Away from oil, towards new energy systems. This is a gigantic task, and could trigger a whole new era of huge economical growth. 2012 would be a nice date to start this, the oil price shock of 2008 laid the psychological foundation. Until then, we should expect some more nasty surprises and frustrating markets.

  2. 2 bill rook

    hi does the use of an exponential moving average set make alot of difference?

  3. 3 Jeffrey Lin

    TR- Using any indicator including moving avgs to the extreme of depending on these as your investment thesis is dangerous. I think its more important for traders to understand what a particular indicator is measuring. in this instance rising moving avgs simply mean price has been rising- 20 SMA indicating recent strength 50 SMA indicating intermeidate term has been going well too. anyone looking at the last 3 months chart can see the strength. I myself use a bunch of indicators, but olny as guides. Price action always is the here and now. if u waited till the moving avgs gave u a buy signal you've already missed a chunk of the move. now i might look for a pullback to the 50 SMA for a better entry and it would still fall within Merrill's reseach statistics of Moving Avg cross success!

  4. 4 Babak

    steffan, thanks I was able to get the gist of it through google translate

    bill, yes, most definitely. The stats I cited are run with a simple MA.

    Jeffrey, agreed. This is a way to look back and see how an indicator performed historically. That is one thing, how it will perform going ahead is another.

  5. 5 wayne

    Simple Moving Average Crossover Systems

    Probably the first rule in most stock market trading 101 classes is that the trend is your friend. Given the big moves in commodities in 2008, a lot of trend following systems had a good year. A trend following strategy that is followed by many technicians is the 200 Day Moving Average Crossover Strategy, where one goes long when the index of interest is above its N day moving average and either moves to cash or goes short when it is below the average. Given that hardly a day goes by that you don’t hear an analyst mention the market’s position relative to it’s 200 Day Moving Average, I thought that it would be worth the effort to restudy the significance, or lack thereof, for not only the 200 day, but other moving average trend following systems as well.

    Test Approach:

    1. Evaluate 50, 100, 150, 200, 250, 300 and 350 day moving average cross over strategies for the S&P 500 Index strategy.
    2. The time period that I used was 1970 through June 16, 2009. Moving averages in 1969 were calculated to initialize the first day’s trade.
    3. If you are short, go long the S&P 500 Cash Index at the close of that business day when the S&P 500 closes 1% above the N day moving average. Since this is a longterm trading system, the 1% filter is used to avoid annoying day to day whipsaws.
    4. If you are long, cover and go short the S&P 500 when the S&P 500 closes 1% below its moving average.
    5. Evaluate the performance in terms of average annual percentage return.

    Summary of Moving Average Crossover Study

    Number Average Avg Ann Avg Ann
    Of Days Annual Return Return
    Moving Avg Return On Longs On Shorts
    50 -0.33% 6.14% -9.77%
    100 0.07 6.31 -9.71
    150 3.12 8.78 -6.05
    200 5.43 10.24 -3.32
    250 5.27 9.70 -3.71
    300 5.50 9.60 -3.28
    350 4.38 8.63 -5.08

    As a point of comparison, the average annual return for a buy & hold strategy over this time period was 6%. With dividends included, the return was 9.4%

    Observations

    1. The 200 day moving average crossover rule is the most profitable of the 6 moving averages studied for the time period selected.

    2. None of the Moving Average strategies were profitable on the short side, which is due primarily to the fact that the S&P 500 has a positive bias during this time frame as it does over most periods.

    4. None of the strategies yielded more than the 9.4% return obtained with buy & hold plus dividends.

    Since the short signals were counterproductive, I ran one derivation of the above study for the 200 Day Model, where one goes long the S&P on the up crossover and goes into 3 month Treasury Bills on the short side. That strategy resulted in an 8.664% return, but still less than the return of a buy & hold strategy plus dividends .

    Although the Tbill strategy on the short side returns were very comparable to the return of buy and hold, it had a much better risk adjusted performance, as it spent 1/3rd of trading days with no exposure to the volatility of stocks. Also, the worst year for this trading strategy was 1990 with a 10.56% loss. Important to note, that a passive buy and hold strategy would have resulted in 7 years with worse performance than -10.56 and as you know a 38.5% loss in 2008

    Conclusions

    The 200 Day Moving Average Cross Over trading strategy has some utility as a trend following system, but it is only slightly better than the return of passive buy and hold. However, the strategy has appeal due primarily to the reduction in risk and elimination of large drawdowns, which traders understand impacts their Sharpe Ratio. Although the 200 Day strategy is the best of the 6 moving average strategies studied, I have been taught and have learned through experience that this may not necessarily be the case going forward. My conclusion is that this particular trend following system has some utility in a longterm trading model but is not the net total solution to trend following and should only be used in combination with other forecasting tools.

    [Ed. removed due to formatting error]

  6. 6 wayne

    Since you guys were discussing moving avg trading rules. I posted a study I did this month on such. The original was done in Microsoft Word and the Tabs didn't come across in this links format. If there is interest, I will attempt to clean up and resubmit, or else try to clarify any confusion.

  7. 7 Babak

    Wayne, I sent you an email. Send me the file and I'll put it up for others. That way it is more accessible. I've removed the cut/paste you did because it wasn't formated.

  8. 8 Jerry

    Are these statements contradictory?:

    Delisle adds that a golden cross appears to have more validity when it occurs with a rising 200 day moving average - something we had in 2003 but do not have now.

    In the Merrill Lynch report prepared by Mary Ann Bartels, it continues to distinguish between golden crosses that happen with a downward long term moving average and those when the long term moving average is rising: Of the 42 Golden Cross signals triggered since 1928, 20 have occurred with the 200-day moving average in a declining trend or lower than it was 30 trading sessions ago. These signals on average have generated 12-month returns of 13.3%. The remaining 22 signals occurred when the 200-day moving average was rising or higher than it was 30 trading sessions ago. The returns for these signals were much lower and on average generated 12-month returns of 5.7%.

  9. 9 Babak

    Jerry, they are contradictory but because Bartels looks at golden crosses going back to the 1920's while Delisle looks at only 14 recent ones.

  10. 10 Dave

    And that would be the same Bartels (look at a daily chart while viewing).

    Tues. May 6 2008
    "Hedge funds are bearish, and it could make sense to bet against them and go long on stocks, says Mary Ann Bartels."

    Twice in 2008, she stated on CNBC that we were about to see major rallies just before the mkt fell out of bed.

    Of course, anyone can be wrong, but what gets me is that she never addressed such colossal mistakes with even an "Oops !". And of course, CNBC would never bring up the faux pas on subsequent appearances.

    As the year progressed it became evident that she had a huge misunderstanding of why the hedgies were sitting on all that cash. They understood that they were going to have redemption problems.

    Btw, i have always assumed that she was the daughter of Bernadette Bartels from the late Wall St Week. So i got a kick out of Bob Pisani saying to her on another video "Of course, you're not a technician." That would be news to Bernadette.

    She certainly is not a worthy successor to Rich McCabe & Bob Farrell.

    Regards,
    dave

  11. 11 Babak

    Dave, thanks for the video. I tend to think of hedge funds as the 'smart money' so I'm not so sure I would have bet against them, except maybe at a crazy point of extremeness. Not putting analysts feet to the fire is one shortcoming of CNBC among many. The Nightly Business Report is the only show that I know which does go back and give updates on previous prognostications when they bring back and analyst or talking head.

  12. 12 Babak

    I've uploaded the moving average study from Wayne. Take a look.

  13. 13 Stefan

    I prefer to look at the slope of one moving average instead of a crossover from another MA or an index itself. Looking for a changed slope of the MA removes some of the fake signals and whipsaws that occur from crossovers. I made two simple charts a while ago showing how monthly MA12 has performed. I´m sorry I don´t know how to make them clickable but just copy and paste in your browser.

    First one shows S&P 500 and using monthly MA12 (it´s like daily MA200) to identify the overall trend, blue=bull, red=bear

    The second chart shows an equity curve (blue) for the MA12 and a comparison to buy-and-hold (red).

    The obvious reason to keep track of a longer MA like this is that you can avoid those devastating drawdrawns that occur once in a while, like 1970ies, 2000-2003 and 2007-2009. Perhaps the next big crash is many years away, but you never know. Another reason is to stay on the right side when trading short term. All dips should be bought when MA200 is trending upwards etc.

  14. 14 Babak

    Stefan, that looks really good. Have you tried it on other indexes? Why not add a short component to it? That is go short when the slope changes. It reminds me of maoxian's 'fuzzy' trend trading system.

  15. 15 Stefan

    Babak: Yes I´m using it on the main index here in Sweden, mostly as a filter to decide between sell or buy signals in shorter time frames. A short component is a good complement for periods like the one we have just experienced between 2007-200x, but I haven´t added it in any charts, but that would be interesting.

    I checked out Maoxian´s site a bit, do you know if he´s explained the "fuzzy" system somewhere?

  16. 16 Babak

    Stefan, no, he hasn't as it is 'proprietary' :) but you're welcome to ask him, he's a nice enough bloke.

Leave a Reply



Warning: include_once(sk2_functions.php) [function.include-once]: failed to open stream: No such file or directory in /home/traders/public_html/wp-content/plugins/SK2/sk2_util_class.php on line 2

Warning: include_once() [function.include]: Failed opening 'sk2_functions.php' for inclusion (include_path='.:/usr/lib/php:/usr/local/lib/php:.:.') in /home/traders/public_html/wp-content/plugins/SK2/sk2_util_class.php on line 2

Warning: require_once(/home/traders/public_html/wp-content/plugins/SK2/sk2_functions.php) [function.require-once]: failed to open stream: No such file or directory in /home/traders/public_html/wp-content/plugins/SK2/sk2_core_class.php on line 31

Fatal error: require_once() [function.require]: Failed opening required '/home/traders/public_html/wp-content/plugins/SK2/sk2_functions.php' (include_path='.:/usr/lib/php:/usr/local/lib/php:.:.') in /home/traders/public_html/wp-content/plugins/SK2/sk2_core_class.php on line 31