Over the past three years, I have often written about Fibonacci analysis or included an interpretation of it within other articles. The stimulus for this article was my discussion with a fellow analyst about the recent action in gold, which appears to have completed its correction from the 2008 highs. If this is the case, then in order to determine potential upside targets, I updated my Fibonacci analysis and thought a refresher course and subsequent analysis of some current markets might be beneficial.
In my opinion, Fibonacci analysis can be very useful in determining support and resistance levels, as well as for identifying longer-term Fibonacci targets that sometimes take several years to achieve. From a trader's perspective, if you are organized and disciplined, these targets can be useful. However, if you are not, their value may be somewhat limited. Investors will find them quite useful when markets are in a long-term uptrend or downtrend, as they can help determine exit or entry points.
Many traders rely on Fibonacci projection analysis to determine profit taking levels regardless of whether they are trading based on daily or intra-day data. This can be helpful in maintaining discipline. Though this is just one tool, the formation of a bearish candle at a projected high could be a further reason to sell. Conversely, if a bullish candle forms at an important support or projection level, it may support your decision to close out shorts or to buy.
For those of you who want to learn from a Fibonacci professional, I can recommend the book of an old friend, Joe DiNapoli, who wrote Trading with DiNapoli Levels: The Practical Application of Fibonacci Analysis to Investment Markets. Just last year, I reviewed another excellent book, Fibonacci Trading: How to Master the Time and Price Advantage (read my review here) by Carolyn Boroden, and Robert Miner recently released a book on Fibonacci that I am looking forward to reading.
Figure 1
For the first example, I ventured back into history by looking at the Dow Industrials as a way to demonstrate that these techniques have worked for many years and are still working across many different time frames. This quarterly chart highlights the decline from the 1929 high of 381 to the 1932 low of 41. This period was dramatic, as the candle chart clearly displays the series of consecutive lower quarterly closes. The rally from the 1932 lows peaked in 1937 at 194.4 (point a), a 45% retracement of the entire decline. The decline into the 1938 and 1942 lows (point b) held the 61.8% support level of the rally from the 1932 lows. For my projection analysis, I will use the 1932 high at 194 (point a) and the 1942 low at 92 (point b).
The 1932 highs were finally surpassed in the second quarter of 1946, but then stocks retreated sharply in the next quarter and moved sideways for the next three years. One of the basic tenets of Fibonacci analysis is that once that one level of support or resistance is overcome, the next Fibonacci level becomes the next target. In the summer of 1950, the 127.2% projection at 222.25 was exceeded. The calculation is as follows:
1937 High = 194.4
1942 Low = 92
Difference = 194.4-92 = 102.2
127.2% projection = 102.2 x 1.272 + 92 = 222.25
A year later, the 161.8% level at 256.4 was also surpassed, making our next upside target 357, which was the 261.8% projection level. The Dow moved sharply through this level in the last quarter of 1954, and I think the fact that this area offered little resistance is important. In my experience, when a significant Fibonacci resistance or support level is easily overcome, it is a sign of great strength or severe weakness. The 461.8% projection target at 521 was hit exactly during the second quarter of 1956 (point c), and tested again the next quarter before a two-year correction. It is interesting to note that this correction just retraced 38.2% of the previous rally.
Let's look at something a bit more recent in the MSCI Emerging Market ETF (EEM) from 2003 through early 2008.
|pagebreak|
Figure 2
The correction from the April 2004 highs to the May lows looks fairly small on the chart, but was actually a 25% decline. EEM edged higher from the lows for the next few months and surpassed the April highs in early November.
The next upside target was the 127.2% target at $64.30, which was surpassed just a few weeks later with the 161.8% target hit in mid-February of 2005. EEM went a bit higher for a couple of weeks before starting a seven-week correction. It was not until September (point 1) that the 261.8% level at $82.05 was hit, and EEM just made it to the $85 level before reversing. The pullback from these highs was brief, and by April 2006, the 423.6% target at $103.30 was reached. The 27% drop from the May highs took many out of the market, and I have found the pattern of this correction to be one of the best for making Fibonacci projections.
Over the next few months, EEM moved gradually higher, and in October (point 2), the 61.8% resistance was overcome, suggesting that the uptrend had resumed. The week of February 8th, EEM hit a high of $119.58, just surpassing the 127.2% target at $119.38. The sharp drop in late February 2007, fueled by concerns over the Chinese market, caught many by surprise as the selling spread to the other emerging markets. From a Fibonacci perspective, it was not that dramatic, as EEM barely exceeded the 38.2% support level.
In just four weeks after the lows, EEM was once again making new highs. The 161.8% upside target at $129.72 was overcome by mid-June. The sharp plunge in August slightly violated the 50% support level of the rally from the February 2006 lows. however, by October, as the major US markets were topping out, the 261.8% target at $159.62 was overcome. EEM made its eventual high at $167.48. For those of you who have not considered using Fibonacci analysis, I hope this will wet your appetite. Now let's look at some current markets.
Figure 3
First, let's take a look at the price activity in gold that lead up to the high of $1033.90 in 2008. The sharp May to June correction in 2006 was just the form I find ideal to make projections. The weekly triangle formation (blue lines) was completed in September 2007, and the 127.2% projection at $782 was hit by the end of October. Gold continued to move higher until it reached the 161.8% projection at $846 in November, where the rally stalled. The correction was brief, however, as gold moved higher into 2008, reaching a high of $1033.90, very close to the 261.8% projection at $1032.
Now let's look at the current picture.
|pagebreak|
Figure 4
The decline from the 2008 highs found support at the peaks from 2006-2007 in the $680-$700 area. The close on January 23, 2009, above the 61.8% resistance level, suggests that the correction is over. Once above the 2008 highs, the first upside target is at $1130 (127.2%) with the 161.8% target at $1252. If this level is overcome, then we have the 423.8% target from the 2006 correction (point a-b) at $1332. If the April gold can surpass the $950 level, then the upside targets at $1130 and then $1252-1332 look likely. Once above $1332, the 261.8% target from the 2008 correction stands at $1604.
Figure 5
How about the Dow Industrials? Using the January 25, 2008 lows (point a) and the May highs (point b), the first downside target at 11,200 was hit at the July lows. The bear market rally from the July lows (point c) carried into mid-August, just exceeding the 38.2% resistance (point d). After the decline resumed, the 161.8% target at 10,700 was hit the week of September 19th. The selling picked up in October, as the 261.8% level at 9150 was easily broken. The November lows at 7449 coincided nicely with the 423.6% projection using the swing from the lows (point c), to the highs (point d). If this level is broken, the next downside target is at 6700, which is the 423.6% projection using the a-b rally.
Figure 6
The interest rate market has been wild for the past year, but this is often the case during bear markets. The long-term chart of the two-year T-note futures shows that they bottomed in May 2000 (point 1) just after the Nasdaq peaked, and then rose steadily for the next three years. They peaked in June of 2003 (point 2), three months after the stock market had bottomed. The correction from November 2001 (point a) to March 2002 (point b) was used for our Fibonacci projections. Incidentally, this correction just barely exceeded the 38.2% support level of the rally from the 2000 lows. The 127.2% projection at 106.92 was the first upside target, and it was hit the week of August 9, 2002, as the high was 107.03. The T-notes continued higher for the next nine months, and reached the 161.8% target of 107.94. The final high for this rally from the 2000 lows was at 108.62. The T-notes, after a multi-year decline, finally bottomed in June 2007, as the first cracks in the financial system began to appear. For our current projection analysis, we will use the correction from the March 2008 highs (point c) to the June 2008 lows (point d). By October 2008, the prior highs had been exceeded, giving us the next upside target at 108.93 (127.2%), which was exceeded at the end of November 2008. So far, the high has been at 109.51, with the 161.8% target at 110.12, which could easily be met if stocks make further new lows. The 261.8% target stands at 113.58.
Even though there is no room for charts, I have a couple of comments on other markets. Copper has held just above the 161.8% downside projection at 125.50, and has bounced nicely. This could be the start of a bottoming process, but it could take some time. Soybeans gapped below their 161.8% downside target at 811 before reversing back to the upside, so this is another market that should be monitored. In the next article, I will take a historical and current look at currencies and in particular the Euro.
