Andrew Coles, 7 April, 2011
Summary of post
One observation in my recent posts, especially March 23, has been the relative strength of the US markets and the FTSE and DAX in relation to the so-called PIIGS (Portugal, Italy, Ireland, Greece, Spain) indices and the Euro Stoxx 50 in relation to the main MIDAS S1 subprime curve. Recently, this theme has also been developed by Elliott Wave International in relation to their ongoing wavecounts. To the list comprising the PIIGS and the Euro Stoxx 50, we can also add the Shanghai SSE. At some stage soon, it’ll be appropriate to do a full MIDAS analysis of these indices, especially the Shanghai SSE and the Euro Stoxx 50. Today however I’ll concentrate on the FTSE 100 as a balance to David’s ongoing analysis of the S&P 500.
As always in my posts when discussing relative MIDAS curves and trend sizes, I follow Martin Pring’s conventions in Technical Analysis Explained. Since this book is on the IFTA diploma curriculum, I assume these conventions are technical analysis industry standards. At least they seemed to be when I did the diploma:
Secular trend = 10 years to as long as 25 years
Primary trend = nine months to two years (associated with unfolding fundamentals and the business cycle)
Intermediate trend = 6 weeks to 9 months, though sometimes longer
Short term trend = 2 to 4 weeks (occasionally shorter, occasionally longer)
In Chapter 3 of the book I’ve added shorter-term trends of relevance to day trading.
In what follows, I’ll start with secular term analysis and end with a perspective on the current intermediate trend.
Secular term curves on the FTSE 100
Chart 1 below is a monthly chart of the FTSE 100 from the beginning of 1984 with volume in the lower pane and a 25 month moving average of volume.
Chart 1: active secular degree curves from the 1980s
The chart is a Yahoo finance cash chart, with volume not being recorded until the beginning of 2003. The only three active secular curves on this timeframe are S1, S2, and S3. S1 is a nominal curve, ie, one computed with constant volume to take account of the absence of volume prior to 2003. (I’ll write a separate tutorial post on nominal curves in the near future.) S1 is also a Hawkins calibrated curve (see Chapter 9 of the book). As we can see, S1 caught the bottom of the 1987 crash, and since then has actively supported the 2003 and 2009 lows, albeit in the latter case with some price porosity.
I’ve left S2, another nominal curve, on the chart because it too supports the 2003 and 2009 lows, albeit with deeper levels of porosity.
S3, the third nominal curve, supported the major 2001 pullback and was also active in 2010 in halting the July low.
Of course, after S3's launch in 1993 many more secular degree curves were active on this index. However, none are presently influential so they’ve been removed.
A final feature of Chart 1 is the possibility, in Elliott terms, of a descending triangle, with a putative wave d currently in progress targeting the 6,500 level. A wave e to complete the triangle would be consistent with secular degree wavecounts on other international indices that anticipate a wave c in a standard ABC zigzag formation. I’ve sketched this tentative triangle pattern on this chart.
The secular term MIDAS Displacement Channel
Chart 2 below is another Monthly chart with S2 and S3 removed. The feint lower curve from the left is the 1984 S1 calibrated curve. This chart displays my MIDAS Displacement Channel (NYSE:MDC), which was the subject of an article in Stocks & Commodities magazine and is discussed in Chapter 14 of the book.
Chart 2 with the MIDAS Displacement Channel
Here, the MDC is also a nominal (constant volume) version, and is also of secular proportions. The lower boundary of the channel fitted at the green arrow at a 34% displacement captured the 2009 bottom much more accurately than the 1984 S1 calibrated curve, and is now obviously a secular degree support area below the 1984 S1 curve. The upper boundary of the channel was fitted at a displacement of 28% at the green arrow and is a major secular resistance curve, currently at the 6,700 level. Since MIDAS curves will sometimes repel price when there’s a small amount of chart whitespace intervening (a phenomenon I call “suspension”), the 6,700 level is consistent on this secular degree scale with 6,500, which was suggested above in relation to the possibility that the FTSE 100 is charting out a secular degree descending triangle.
Primary degree curves on the FTSE 100
Chart 3 below is a weekly chart. Starting from the bottom of the chart, the lowest feint curve is the 1984 calibrated S1 curve. Above that I have the first primary degree curve (red). This is the main subprime curve I’ve discussed in several earlier posts, notably in theMarch 23 entry.
Chart 3: with Elliott Wave count and S2 MDC and S3 MDC
Primary curve S2 is actually another MIDAS Displacement Channel (MDC). The MDC was designed for sideways markets, where Levine’s original curves were ineffective; however, the MDC also works extremely well in trends that aren’t accelerating rapidly, where the upper channel will catch the highs in uptrends and the lower channel will capture the lows in downtrends. We see this here.
So far as the S2 MDC is concerned, it was fitted at the olive arrow at a 10% displacement and captured the Elliott waves 3 and 5 of the first impulse. It has had little effect in halting the new price rise since mid-March.
S3 is also another MDC, this time fitted at the blue arrow high (subwave 3 of wave 3) at a displacement of 9%. The standard S3 MIDAS curve below the upper channel caught the wave 2 and wave 4 bottoms of this second Elliott impulse, albeit with some porosity. The 9% upper channel caught the recent mid-February wave 3 top and is now proximate resistance to the upside since mid-March at the 6,200 level.
6,200, then, is the first intermediate trend target on this chart. However, 6,250 marks the first impulse (= 1.00) multiplied by .618, so the proximity of the Elliott/Fibonacci target plus the MDC target should be taken seriously.
An intermediate degree Topfinder (NYSEMKT:TF) on the FTSE 100
Chart 4 below is a daily chart of the FTSE 100, this time in Equivolume. Equivolume charting dispenses with time along the horizontal x axis in favour of volume; as a result, it’s possible to set an extrapolated price target alongside the firm cumulative volume prediction of the Topfinder or Bottomfinder.
Chart 4: TF measuring the intermediate trend
To get our bearings, the swing low on this chart is the mid-March low. Since then, the price upside has been fairly parabolic. This means that the new intermediate trend S1 (blue) curve has already displaced from the price uptrend, thus setting the condition for the launch of a Topfinder from the mid-March low.
David mentioned in his post of a few days back that currently it’s not easy to fit a TF to the S&P 500 because the upside since mid-March has been so relenting. I agree, but it is possible (just) to fit a TF to the lower right of the Equivolume bar highlighted by the grey arrow on this chart. This fitting is giving a D reading of 190,000,005 shares. This is the predicted amount of cumulative volume that must be entirely burned before the TF completes, and hence the firm cumulative volume prediction associated with this phase of the trend ending.
At the point of fitting, D is 53% complete, meaning that there’s another 47% of D still to run. By using Equivolume charts I can mark a vertical (dotted) line on the chart where the cumulative volume in D will expire. I can also create a linear regression slope through the price upside since mid-March in order to meet the vertical cumulative volume prediction. Where they intersect is the estimated price target. We see this in profile most clearly with the magenta backwards shaped “L”, which was an iconic feature of Levine’s original Topfinder/Bottomfinder.
Whereas the cumulative volume number is the firm prediction in MIDAS theory, the time target is only an extrapolated estimate. This is because the price trend could become more or less parabolic as the remaining cumulative volume is used up. If it becomes more parabolic, we’ll obviously get a higher price target; if it becomes less, we’ll get a lower one. At the present time, the estimated price target is 6,290, though it must be reemphasized that this is an extrapolation. The accurate aspect – provided the fitting itself is correct – is the cumulative volume prediction.
Hence, the variance around 6,290 is consistent with the upper channel of the S3 MDC and the Fibonacci projection given a zone of between 6,200 and 6,250. This of course is for the current end of the intermediate trend. However, it could mark the end of the primary trend, albeit the higher primary trend target is identified by the secular degree MIDAS Displacement Channel at around 6,700 (with the putative triangle at 6,500).
A time target for the FTSE 100
Chart 5 is the final chart here of Wilder’s/Sloman’s Long Term Delta (LTD), which is consistent with the intermediate trend. Pivot 24 is due in June (broken line), though again there is some time variance between each pivot, making this date a reasonably accurate forecast within a deviation of a week or two either side of the pivot.
Chart 5: Long Term Delta – June = broken line
June of course coincides with the end of QE2. I’m not going to waste time discussing QE2 here, as the topic is viral in the financial blog world. I’ll just add that June was also the date for the next S&P 500 pivot, according to my solution to the LTD on that index. The S&P 500 LTD called the mid-February top accurately, but in my February post I was a little premature in assuming that the LTD on this index would not invert. It did invert, as the upside since mid-March has shown. This means that June is a time target now for the upside rather than the downside. Fortunately, inversions occur only at the end/beginning of each Delta cycle, so the short term ambiguity exists only once in every four years on the Long Term Delta.
Intermediate and Primary degree trend MIDAS price (and LTD time) targets as above.
As an afterthought, I’m throwing in Chart 6 which highlights a primary degree negative divergence on the MACD. One obviously has June in mind for this, and it’s consistent with divergences I’ve highlighted in earlier posts on the Baltic Dry Index and the VIX vis-a-vis subprime S1 MIDAS curves, and the short net positioning of the funds and commercials in the COT Report on the S&P 500.