Archive for June, 2011

I will not do a mid month review of the S&P 500 this month since I am away on vacation.

David G. Hawkins

Tweet This Post

What follows is an article to appear in the July 2011 issue of Forex Journal (http://www.forexjournal.com/). As a result, this is a much longer post than usual. The Forex Journal has extremely good content for forex traders (cash and futures) and readers are highly recommended to request a trial issue at the website.

I’ve posted this discussion because a number of purchasers of the book have expressed an interest in seeing more examples of nominal/constant volume (second generation) MIDAS curves, the MIDAS Displacement Channel (MDC), and third generation curves. The following discussion mainly focuses on the cash EUR/USD charts and so contains many nominal curves. The MDC also uses nominal volume data. Third generation curves – ie, curves that remove the “AP” in the VWAP and replace it with other fractal data sets – are here plotted on the spread between Eurex 3 month EURIBOR futures and CME EURODOLLAR futures (Figure 7). For more on third generation curves see Chapter 16 and my two articles in the June and July issues of Active Trader.

Other analysts don’t seem to have picked this up, but the weekly chart of EUR/USD is sporting a large primary degree Gartley 222 pattern. Gartley 222s are rare patterns of some distinction when they form because of their exacting Fibonacci relationships (Figure 5).

There’s also a brief discussion of gold (Figure 2) but I have a separate MIDAS post on gold I’ll be posting in the next couple of days.

A.Coles, June 13th 2011

*

In the December 2010 issue of this journal, I discussed one of our new innovations in the MIDAS approach that allowed it to be applied to the volumeless cash FX markets, despite the MIDAS system being rooted in the formula for the volume weighted average price (VWAP). This review of EUR/USD is possible in virtue of this innovation.

Briefly, we replace market volume in the MIDAS formula with constant volume. More precisely, one unit of cumulative volume inserted into the volumeless MIDAS algorithm ensures that there’s a nominal volume weighting in the MIDAS curves. As I demonstrated in the December article, the result is a highly effective application of MIDAS curves to the cash FX markets (see also Chapter 10), albeit advanced applications of MIDAS curves involve rule-based applications of both nominal and standard (market volume) weighted curves (Chapter 11).

,

Secular degree MIDAS Support and Resistance Levels

I begin this mid-term technical review of EUR/USD by taking advantage of this innovation and applying nominal curves to the very longest secular-term trends influencing the pair in the weekly data in Figure 1.

Figure 1: weekly chart of EUR/USD with secular degree nominal (constant volume) curves

(source: eSignal and Metastock. www.esignal.com and www.equis.com)

Despite the launch of the euro in January 1999, the secular trend (10 to 25 years) can be analysed because data suppliers such as eSignal add pre-1999 data in the form of the European Currency Unit and European Unit of Account, which is based on the Deutschemark.

Moving from left to right, the only pre-1999 secular MIDAS curve of relevance to current price activity is the one launched from the secular degree 1979 top (green), which ended a 9 year uptrend in the Deutsemark. This curve was last active in mid-1997 (point (1)) and is now a major floor at the 1.10 level. R1, from another secular degree top in early 1995, played a major resistance role in mid-1998 (point (2)). Since then, it has had a highly significant role in halting the euro’s decline in the early months of the Eurozone sovereign debt crisis at the start of 2010 (point (5)).

S1 is a MIDAS indicator I developed called the MIDAS Displacement Channel (MDC, for short). The MDC can be created from nominal and standard curves. Its main role is in sideways markets, where the original MIDAS curves tend to move to the centre of price activity. However, the MDC can also perform very effectively when capturing the highs in uptrends and the lows in downtrends. The indicator consists of the original MIDAS curve plus two (or more) additional curves that are displaced from it on either side by a user-determined percentage. The percentage is calculated by fitting the channel to the first higher high or lower low beyond the original MIDAS launch point. In Figure 1 the upper band is fitted to the first significant swing high at the blue arrow at a displacement of 19%, while the lower is fitted to the Greek sovereign debt low at point (5). As we see, the standard MIDAS curve played a major support role in mid-2005 at point (3), while also dramatically halting the sharp post-Lehman decline in mid-2008 at point (4). The upper curve has been equally effective as a resistance band, halting the uptrend twice in late 2003 and late 2004 and then halting the sharp uptrends in the euro’s current congestion at points (6) and (7).

Major secular degree support levels in Figure 1 are:

• 1.28 = S1, original curve
• 1.20 = R1 (inverted from resistance to support) and the lower 6% displaced band of the MDC
• 1.10 = 1979 curve (green)

Major secular resistance level:

• 1.53 = upper band of MDC, displaced by 19%

.

Primary Degree MIDAS Support and Resistance Levels

The primary trend (9 months to 2 years) is currently of keen focus for most traders of EUR/USD, with interest rate expectations on the one side evenly matched by concerns affecting broader risk sentiment.

Euro interest rate expectations in the 3 month EURIBOR are tightly correlated with movements in the euro, and the bottoming of 3m EURIBOR rates at the start of 2010 (see Figure 7) reflected core Eurozone fundamentals in Q1 of 2011 (recently confirmed at 0.8 percent GDP, seasonally-adjusted, 2.5 percent). This resulted in the first rate raise in April for three years to 1.25 percent. Today (June 9) the ECB deferred a further rise, but the post-meeting press conference strongly hinted at a further July increase and possibly beyond, reflecting ongoing inflationary concerns above the ECB target of 2 percent. These expectations have been mirrored too in the widening spread between the 3 month EURIBOR rate and USD counterparts such as the 3 month Eurodollar rate (Figure 7).

The flipside to these interest rate expectations is widening peripheral spreads in the Eurozone, exacerbated by recent market scepticism over the need to restructure Greek debt plus Moody’s latest downgrade of its credit rating and poor Greek GDP numbers. The ECB’s government debt-buying programme isn’t intended as a form of US-style quantitative easing and so isn’t so directly implicated with inflation in virtue of the ECB’s sterilization programme; in practice, however, the latter has consistently failed over the past year, so increasing peripheral spreads and the cost of insuring peripheral debt in Credit Default Swaps is impacting bond market sentiment and short-term inflationary concerns in the euro. Recent historical data have shown that the intermediate trend in the euro is positively correlated with falling peripheral Eurozone spreads and declining costs in the CDS markets. Thus, rising peripheral spreads and the cost of insuring peripheral zone debt in the CDS markets in comparison with rising 3 month EURIBOR in the futures markets creates a stand-off between normally inversely correlated markets and it has potentially powerful implications for the euro.

As far as risk appetite is concerned, up to the beginning of June the US dollar’s correlation with equities has been at its highest on record. However, this correlation has recently weakened and gold too has lost its momentum. Suddenly the relation between the dollar and risk markets has become harder to determine and various scenarios are possible.

One possibility – depending partly on the Fed extending its bias beyond June – is that loose monetary policy will reinvigorate the correlation between the dollar and risk markets, partly because assets such as gold are seen as hedges against dollar weakness, and partly because the dollar will continue its role as carry for risky assets. If, according to the current wavecount on US equities, there is soon to be a more prolonged bout of equity weakness affecting broader commodity risk (perhaps triggered by a further tipping point in US economic performance such as May’s employment report) the opposite will occur, and the deleveraging will herald a sharper and more prolonged depression in gold and other commodity prices. On this scenario, the catalyst for improving dollar fortunes won’t be yield spreads but a more prolonged interregnum in risk appetite. However, this scenario could usher in a bout of gold buying at the expense of the dollar, since gold is sometimes seen as a store of value in times of economic underperformance. It mustn’t be forgotten either that yield momentum as a result of the termination of the QE program could also invigorate the dollar, as certain proponents of the dollar/gold cost of carry argument observe.

Technically, momentum on the US dollar index is faintly supporting greater dollar upside. Moreover, I’m looking at the primary trend in gold from the October 2008 low with considerable interest insofar as a Topfinder of primary degree recently completed four weekly bars from the May 2 top. Since then, rising gold prices have slowed dramatically, suggesting that the current move up may be corrective. Alternatively, I have another Topfinder fitted fractionally below the previous one indicating that there’s a mere 10.7% cumulative volume left to run on this trend. Either way, as Figure 2 indicates, the Topfinders are suggesting extreme caution when it comes to further bets on dollar weakness.

Figure 2: Weekly chart of euro continuous futures with the MIDAS Topfinder with a mere 11.3% cumulative volume left to run.

(source: eSignal and Metastock. www.esignal.com and www.equis.com)

As regards MIDAS support and resistance levels on the primary trend, Figure 3 is another weekly chart covering the period between the 2008 top (green arrow) and the current level of 1.40 in early June 2011.

Figure 3: weekly chart involving primary degree analysis of EUR/USD with a smaller scale MIDAS Displacement Channel

(source: eSignal and Metastock. www.esignal.com and www.equis.com)

The upper band of the MDC is displaced at the black arrow at 8% and perfectly captures the 6 May 2011 high at 1.50 (point (1)). The lower band was displaced at the black arrow at 13% and again robustly captures the Greek sovereign debt low at point (2). Primary degree support and resistance levels from this indicator are:

• resistance = 1.50 = upper displaced band of MDC
• support = 1.20 = lower displaced band of MDC (see also secular degree support in Figure 1)

.

Upside Price Targets on the Primary Degree Gartley 222 Pattern

Before proceeding to the intermediate trend in Figure 6, it’s worth drawing attention to the primary degree Gartley 222 pattern on the euro in Figure 5, which in Figure 3 was the subject of the primary degree MDC. Genuine Gartley patterns are rare on this scale and are a wonder to behold because of their exacting Fibonacci relationships. This pattern does not seem to have been recognized by other analysts, so I highlight it here with its forecasting implications.

As in Figure 4, the Gartley 222 pattern consists of a prior impulse move (XA) followed by a pullback in two shorter waves (AB, CD), where CD approximates AB in time as well as in a price-based “measured move”. Point D is ideally reached within a Fibonacci .62 to .79 correction of the XA move. BC approximates AB * .38 to .79, while CD approximates BC * 1.27 to 1.62. In the bullish version of the Gartley 222, D will typically terminate at a Fibonacci correction of .62 to .79 of the entire XA move.

The proximate upside price target of the bullish Gartley 222 is the length of the wave BC, which coincides with the red dotted resistance line in Figure 5. If this price objective is met, there are Fibonacci extensions of wave A-D of 1.23, 1.38, 1.50, and 1.61.

Figure 4: Bullish Gartley 222 with Fibonacci ratios

As we can see from Figure 5, the Gartley 222 on the euro has completed alongside the first upside target, the length of BC. Higher targets on the euro futures would be:

• AD * 1.23 = 1.69
• AD * 1.38 = 1.75.5
• AD * 1.50 = 1.80.5
• AD * 1.61 = 1.85

These are targets for the upside primary trend if indeed the current pullback is a paring of recent gains as opposed to a trend reversal.

Figure 5: weekly chart of continuous Globex euro futures with Gartley 222 pattern of primary degree proportions.

(source: eSignal and Metastock. www.esignal.com and www.equis.com)

.

Intermediate Degree MIDAS Analysis

The intermediate trend lasts between 6 weeks to 9 months. It can be seen running in Figure 3 between the low at point (2) and the high at point (1). We’re now starting a second intermediate trend.

Figure 6 is a daily chart of EUR/USD, still using volumeless cash data and not futures. At this point, I should say that when it comes to futures FX data, a MIDAS analysis can use Globex volume or open interest to create OIWAP curves. An OIWAP curve is a second generation MIDAS curve alongside nominal curves. However, here I’ll stick with the volumeless cash FX markets.

Figure 6: weekly chart of euro continuous futures and intermediate degree curves

(source: eSignal and Metastock. www.esignal.com and www.equis.com)

From left to right, we see R1 (red) from the November 2009 top act as major resistance at the two red curves, while R2 (blue) has since August 2010 switched its role from resistance to support, and is now closely pacing S1 (green).

For the short term trend (duration of between 2 and 4 weeks) in the euro, we see that price has been critically locked between S2 (olive), launched from late December 2010, and the new R1** (orange) launched from the recent mid-April high.

At the time of writing, S2 has held sway over R1**, in which case (in MIDAS terms) we say that the uptrend is still intact. Accordingly, a new support curve, S3 (black), has just been launched. It will now offer critical support in this new intermediate degree uptrend, particularly for the resolution of the upside price targets in the Gartley 222 pattern, since the recent support at S2 is also bolstered by the extension of the AC trendline in the latter.

• Critical intermediate degree support = 1.34 to 1.35.5 = R1, R2, and S1
• Critical short term support = 1.40 = S2 and the new S3

• Critical short term resistance: none, since R1** has recently been broken and we look to the previous price high from where R1** was launched

,

Non-Priced based Support and Resistance with Third Generation MIDAS Curves

More recently, I’ve created third generation MIDAS curves, which involve replacing the average price in the original VWAP-based MIDAS formula with other time series displaying fractal trend characteristics related to, but sufficiently different from, price. This sameness and difference means that the curves identify many additional market inflection points that cannot be located with price-based curves. Although third generation curves don’t produce priced-based support and resistance, they do create robust signals that either coincide powerfully with price-based targets or expose hidden market inflection points. Consequently, they represent an important extension of MIDAS studies, giving rise in particular to two new MIDAS setups I’ve called Dipper and Inversion (for the former, see Chapter 16, p377).

I’ll complete this review of the euro by looking at EUR/USD interest rate differentials.

Interest rate differentials via yield or bond spreads tend to work best with regard to the primary and secular degree trends. However, their fractal trending means that they’re another ideal dataset for third generation curves.

Figure 7 contains a weekly plot of the euro in the lower pane and above it an inverted plot of the spread between the EUREX 3 month EURIBOR continuous futures and continuous Eurodollar futures. It’s also possible to use the inverted spread between the EUREX Schatz and US 2 year T note futures or EUREX Bund and US 10 year T note futures. However, third generation curves seem to perform best on the short end of the yield curve. The underlying of the EURIBOR futures are time deposits denominated in US dollars at banks outside of the USA, while the CME futures are determined by the market’s expectation of the 3-month USD LIBOR interest rate expected to prevail on the settlement date.

Figure 7: weekly chart of EUR/USD with an inverted spread of the 3 month Euribor and Eurodollar continuous futures in the upper pane.

(source: eSignal and Metastock. www.esignal.com and www.equis.com)

The interest rate differential in the upper pane is analysed with the MDC (upper band = 2.75%; lower = 2%) and nominal MIDAS support curves. This combination has produced seven major Inversion signals. The latter are so-called because they identify market tops in uptrends and lows in downtrends. Such inflection points are out of reach of MIDAS curves created on the price plot (albeit the MDC is also extremely good at identifying these points). The most spectacular of these inverted signals (highlighted in the green rectangle) is the 2008 secular degree top in EUR/USD caught by the upper band of the nominal version of the MDC.

Going forward, the outer extremities of the MDC will be of obvious relevance if the yield spread in EUR/USD widens and risk appetite increases correspondingly to fulfil higher price targets in the Gartley 222 pattern. Failing this, the mid-2010 low in the euro is well supported by the middle curve of the MDC plus several more recent curves that are currently supporting the widening spread between European and US yields. Pullbacks in the intermediate and primary degree trends in EUR/USD will be heavily dependent on the curves launched on the spread from the 2010 low. Any break of these third generation curves will target the aforementioned middle curve of the MDC launched from 1999 plus the two other curves clustered around it from the 2002 and 2004 highs.

Tweet This Post

by David G. Hawkins

Here is my monthly update of the daily, weekly and monthly bars charts of the S&P 500.

Short Term – Daily Bars Chart

The first chart here is the daily bars chart.  A lot of interesting things have happened on this chart since I last updated it on May 15th – see my last post here for comparison.  First, price plunged sharply down, hit and bounced up from the curve labeled “Old S1″ which was launched from the March 16th low.  After that, in only two days of strong uprising, it hit and turned down from the new R2 curve, which tells us that the downtrend that started May 2nd is still intact.  Then, in another two days, it went down and formed a bottom at a price level of about 1313, from which it rose sharply.

Now, looking at this chart in my last post here, there is no support curve at the 1313 level.  ”Old S1″ was up at 1319 and didn’t provide support, and the weekly S3 level was far below where price actually found support.  I always try to have set up in advance a ladder of support levels below price so we may watch price find support as it comes down.   But sometimes, price decides to find support at a level that one may not have previously identified on the chart, and that’s what happened here.  Look now at the curve in bold green on today’s updated chart, launched from the Jan. 28th low, one which heretofore I hadn’t kept on the chart.  It provided perfect support for the most recent low in price.  Sometimes this happens, since keeping every support curve on the chart would be too cluttered.

Here’s how to treat this situation in a methodical way.  As soon as it looks like price may be lifting off a previously unidentified support, go back and launch curves from older lows to see if one may be providing support here.  If you cannot find such, then one should not place much significance on the price rise.  However, if you can find a significant support curve that does explain the turning in price, then one would have confidence to base ones trades on such a price rise.

In the past week, price rose smartly from this recent support, blasted right through the new R2, but came to a screeching halt right at R1.  Therefore, this is a crucial point today; if price turns down from here, then the month long downtrend is still in effect; if price definitively breaks and closes above R1, then the downtrend has ended.  Today should be very interesting!

Intermediate Term – Weekly Bars Chart

The second chart here is the weekly bars chart updated through May 31st.  We see that over the last few weeks, price broke but did not close below S4.  Had it closed below S4, I would conclude that the uptrend that started last summer has ended.  But, those two breaks below S4 are a warning that this uptrend is weakening.  I would not at all be surprised to see price come down and test S3.  Both the Money Flow Index (upper pane) and the Volume Weighted MACD (lower pane) continue to show weakness.  The TopFinder is about 52% done, but as I said in my last post, I don’t have much confidence in that one proceeding to its conclusion.

Long Term – Monthly Bars Chart

The third chart here is the updated monthly bars chart.  We see that the most recent two candles are implying a top; and we see that price seems to be turning down from the 76.4% Fib level.  The TopFinder is about 3/4 done, looking like it has only a few more months to go, with its end projected to come at the dashed vertical line.  My opinion is that this TopFinder’s ending will dominate over the one on the weekly bars chart.

Tweet This Post