Shares digital

A lack of volume suggests recent gains in the US index may falter

by Simon Griffin

Dow Jones Industrial Index (DJI)

When last discussed in this column, with the Dow at 12,266, I said a test of support from the May 2006 high at 11,675 was likely and would be key to determining the likely future course for the index. It subsequently hit a low and close of 11,740 on 10 March.

It came close to the previous high and coincided with a test of trendline support drawn off the low closes of April and October 2005 and July 2006, and with the January 2000 high (horizontal black line), so there was much to suggest strong support at this level. As this move occurred, positive momentum divergence appeared and argued in favour of at least a short-term upside correction.

This has also panned out as the index gained to test neckline resistance from the possible inverse head-and-shoulders pattern that has been forming for much of this year. Whether this pattern will work through or fail remains uncertain, however worrying is the lack of volume evident for the recent rise. This suggests the move may fail ahead of a test of the bear trendline drawn off the close highs of last October and December, which comes into play at 12,783 currently.

Realistically I would prefer to see a move above the 200-day average at 13,100 to claim an end to the blood-letting. More likely in its absence seems a further test of support at 11,740 then possibly 11,300. My longer-term suggestion of a possible test of the 50% retracement of the 2003-2007 bull move at 10,725 remains achievable.

This Week’s Hot Chart

Rolls Royce (RR.)

The aero engine maker’s shares were a great play during the bull market of 2003 to 2007 climbing over eight-fold in value. Much of that upmove was dominated by a bull channel, with the shares climbing above the channel return line in early 2006. It seems that recent weakness, which took the shares down from 546p at the end of 2007 to a low so far of 393p in mid-March, has achieved what technical analysts like to see: a proportionate move below the channel line to mirror the previous upside break.

This move also achieved a test of support in the 380p-390p zone that previously supported conclusively in early summer 2006. Volume spiked on the selloff in early February but was significantly more muted during the new low posted in mid-March as positive momentum divergence appeared. Now the bear trendline has been broken to the upside and the shares are pressuring their 50-day average, suggesting gains are to be expected. There is clearly significant congestive resistance to be tackled in the 480p-490p area, with the descending 200-day average currently lurking close to 500p, but a climb back up into the bull channel would see the shares back up at 520p.

Technical talking point

Cycling along in the markets

Markets typically go through cycles of accumulation, uptrend, an unsustainable buying climax that leads to distribution, and a final selloff downtrend before repeating. Observing volume rises can help in detecting moves between phases, but volume is not always available. We therefore need to think about the time nature of cycles. Many economists will be happy with the concept of cycles from a macro-economic standpoint and perhaps until quite recently would have been happy to discuss the concept of a relatively fixed cycle length for economies.

With shares we can use charts to observe cyclic patterns and see where cycle events might occur. The first approach is to take the distance between two obvious turning points (two lows or two highs) and see if, by stepping through the chart by similar periods, it coincides with a turning point. If so, it might have value for predicting events.

A refinement is to use a set arithmetic approach and to take the initial time period of the first cycle and multiply it by the Fibonacci sequence to get a set of cycle lines where time periods gradually increase with each new line. Start overlaying these lines. Some will coincide. Points with several cycle lines close together are more significant than points where lines are isolated.

On the FTSE 100 chart, the 2003 low is taken as the start and for the arithmetic cycle method (mauve vertical lines); the July 2004 low is taken as the next cycle point. I overlay Fibonacci cycle lines (black vertical lines) and set the sequence so its third line coincides with the July 2004 low. The fifth line coincides with the third line of the arithmetic series in early summer last year and very close relatively to the top of the market. It was possible to draw these lines on the chart once the July 2004 low lines had been set, some three years before the chart had reached the 2007 top.