COPPER NEARS RECORD HIGH -- LOGARITHMIC TRENDLINES WORK BETTER ON LONG-TERM CHARTS -- HOUSING INDEX BACKS OFF FROM 200-DAY AVERAGE -- WEEKLY EMA LINES ARE STILL NEGATIVE -- HALLIBURTON ACHIEVES BULLISH BREAKOUT

COPPER NEARS NEW RECORD ... One of our readers asked for an update on copper. The question comes at a good time. Chart 1 shows copper on the verge of reaching a record high over $400. Copper has been trading sideways since the spring of 2006 between $400 and $250. The weekly price pattern also has the look of a bullish "ascending triangle". [The ascending triangle is marked by a flat upper line and a rising lower line]. The relative strength ratio (above chart) measures copper against the CRB Index and shows that copper has been a commodity laggard during those two years. Since the start of the new year, however, copper has started to show more commodity relative strength. That may explain recent buying in copper stocks like Freeport McMoran Copper and Gold.

Chart 1

FREEPORT MCMORAN BREAKS NECKLINE... I'm going to use Freeport McMoran Copper & Gold to answer a couple of reader questions. One question has to do with "inverse head and shoulders" patterns that take place after a correction. It's true that the "best" H&S bottoms take place at the start of new uptrends. That pattern can also take place, however, after downside corrections and is viewed as a continuation pattern. Last Friday I showed FCX breaking through a "neckline" (green line). That pattern has the look of a "continuation head and shoulders" bottom". For that bullish view to remain intact, however, FCX needs to hold over the neckline which should now function as a support line. That may depend on whether or not copper itself hits a new high.

Chart 2

LOG TRENDLINES OFTEN WORK BETTER ... Another question had to do with the choice of a normal price scale versus a log scale in some of my longer range charts. s a rule, log scales are preferred for long-term charts. One of the reasons for that has to with the drawing of trendlines. Chart 3 plots a weekly bar chart of Freeport McMoran (FCX) over the last four years using a normal price scale. As you can see, the rising trendline drawn under the 2005-2006 lows is very flat and hasn't been very useful. Chart 4 plots the same prices on a log scale. [A log scale measures percentage price changes. That accounts for the space between the upper numbers being squeezed]. The same trendline on Chart 4 works very well and helped identify an FCX bottom this January (see arrows). Here's the rule of thumb. A logarithmic trendline will be tested earlier than a normal trendline on a long term chart. That's good for two reasons. It offers a potential support point as in Chart 4. At other times, the break of the log trendline gives an earlier sign of a downside trend reversal. It's usually a good idea to use both scales to see which one is working better.

Chart 3

Chart 4

HOUSING INDEX BACKS OFF FROM 200-DAY AVERAGE ... Another questioner asked if it was reasonable to expect the recent market rebound to continue without upside leadership in financials, retailers, and homebuilders. Arthur Hill showed the first two groups yesterday backing off from overhead resistance. I'll deal with homebuilders. The PHLX Housing Index has been showing better relative strength since January. That can be seen in its rising relative strength ratio (below Chart 5). But the ratio is now stalled at its late-January peak. So is the HGX itself. The index of homebuilding stocks also backed off this week from its 200-day moving average (red line). That puts the homebuilding rebound at an important juncture. In order to signal a more serious upturn, the HGX needs to decisively clear its 2008 highs (and its 200-day line). Failure to do so (especially if that failure is confirmed by new selling in financials and retailers) could doom the stocks market's rally attempt.

Chart 5

BE WARY OF SHORT-TERM BUY SIGNALS... One of our readers correctly pointed out the 13-day exponential moving average (EMA) had crossed over the 34-day EMA on all of the major stock indexes. Chart 6 shows that to be true. That's a short-term positive. What makes me suspicious of the significance of that short-term improvement is Chart 7 which plots the 13- and 34- "week" EMAs. Those more important lines turned negative at the start of the year for the first time in five years. And they're still negative. The black line below Chart 7 plots the difference between the two weekly EMAs. [You can plot that line by plugging 13,34,1 into your weekly MACD indicator]. The spread between the two weekly EMAs is narrowing a bit, but is nowhere near crossing back over the zero line. The rule of thumb is this. "Daily" EMA buy signals work best when the "weekly" EMAs are also on buy signals (from 2003 through 2007). Short-term bullish crossings usually don't carry very far when the weekly lines are negative (as in 2000 through 2002). I'm not saying the short-term buy signal should be ignored. Just keep it in perspective.

Chart 6

Chart 7

HALLIBURTON BREAKS OUT ... Last Friday I showed Halliburton testing the top of a two-year bullish "ascending triangle". Chart 8 shows the oil service leader accomplishing that bullish breakout this week. Arthur Hill wrote yesterday about the problems rising oil prices were causing for transportation stocks. That's the bad news. The good news is that rising oil prices are good for energy stocks.

Chart 8