With the recent action of the Semiconductor Index (up about 12% in 3 days),
it is reasonable for market participants to wonder if "the bottom" is in. After
all stocks have stopped going down, and now that the semiconductors are leading
the charge, well stocks can only go higher.
But I don't buy that view - yet. There is some good technical evidence that
the semiconductors have put in a bottom. The fundamental picture, on the other
hand, suggests otherwise as this would be the first "bottom" in about 5 years
that was not accompanied by a rising yield curve.
More on that below but first the technical evidence.
Anatomy of a Bottom- Technical Evidence
Looking at the price action through my proprietary Price Structure Analysis™ indicator,
we can see that support in the Semiconductor Index was at the 360 level. This
is shown on the weekly chart (figure #1). Support was being tagged at a time
when there was plenty of bearish sentiment towards higher prices in the stock
market (see the sentiment section on the website, thetechnicaltake.com). In
my opinion, this made for a good low risk entry.
Figure #1/ Semiconductor Index/ weekly

Semiconductor sector internals, such as new highs and new lows and advancing
and declining issues, give some credence that this might be a good time to
take a position in the semiconductors as well.
The weekly semiconductor sector chart (figure #2) is shown below. This chart
is comprised of the 15 largest stocks in the semiconductor universe. The top
panel is the price bars; the second panel is the percent of stocks above their
200 day moving average. Ignoring the lines on the graph, this indicator tells
me if I should be playing offense or defense. Being so oversold, it appeared
to me that the semiconductors would be trying to establish a bottom; therefore
this would a good time to go on offense and look to establish a long position.
Figure#2/ Semiconductor Sector Chart/ weekly

The third panel is the trading index (or Arms Index) which is a calculation
that utilizes volume and advancing and declining issues. Essentially, the trading
index tells you where the trading volume is going. Being below the center line,
suggests that the flow of volume is going into advancing issues, and this supported
my notion that the Semiconductor Index was putting in a bottom.
The bottom panel is a McClellan Oscillator of the weekly advancing and declining
issues. I have highlighted on the chart (point 1) a divergence between this
oscillator and the price action. Divergences can be a powerful tool, and they
may pinpoint when the current trend is weakening and when prices may reverse
direction. I have also noted the previous divergence (point 2) that occurred
in October, 2002.
Staying with the technical picture, look at the daily sector chart of the
Semiconductor Index (figure #3). Like all sector charts posted to thetechnicaltake.com,
this chart is constructed utilizing the 15 largest capitalized stocks in a
particular sector or index. The top panel is the price action; the bottom panel
measures the number of new highs and new lows over the past 40 days in this
particular sector. As prices headed lower, the number of stocks in this sector
making new lows was decreasing. This also is a divergence, and one that I find
rather powerful. Even though price is falling in the index, the stocks that
make up that index are doing better on a relative basis. In any case, downside
momentum was decreasing.
Figure #3/ Semiconductor Sector Chart/ daily

In summarizing the technical picture, utilizing the sector charts on the website,
I was able to pinpoint a possible turning point in the Semiconductor Index.
I used the concept of Price Structure Analysis™ to determine support
and my entry point.
Ok that was then and this is now? So tell me why you don't think this move
in the semiconductors will stick?
Anatomy of a Bottom - Fundamental Evidence
These comments refer to the general market. {I will discuss the general market
in greater detail later this week.} The price action has been good and various
levels of resistance have been cleared. Yes, volume has dried up but who is
noticing when prices are moving higher? Despite most of the intermediate term
breadth indicators (see the web site: thetechnicaltake.com) being near their
overbought extremes, prices can move higher for one reason: this rally has
not been embraced by investors.
The chart (figure #4) below shows that the Rydex investors - a group of market
participants that are assumed to be wrong at market turning points – still
have not jumped on board. The top panel is a daily chart of the QQQ, and the
bottom panel is composite indicator looking at the assets in bullish and bearish
Rydex funds. The total assets in all the bullish funds still remain below that
of the total assets in bearish funds. Furthermore, the bullish assets haven't
reached an extreme relative to past levels. All this tells me is that there
are buyers out there. What you should notice from the chart as well is how
the rally that started in March, 2003 was not fully embraced by investors until
almost November, 2003!!!! When this group of investors (ie., the Rydex market
timers) determined it was safe to wade into the market, the party was just
ending.
Figure #4/ Rydex Sentiment/daily

Ok that is the general market. What about the semiconductors? After all it
is hard to envision the market going higher without the Semiconductors leading
the advance.
Is the action in the semiconductors an oversold bounce fueled by short
covering or is it a new round of speculative interest that will propel prices
higher? I believe it is the former because the ratio of long term to short
term interest rates (otherwise known as the yield curve) continues to fall
suggesting a slowing in growth and a slowing in the economy.
The yield curve is an excellent barometer of future expectations in economic
growth. Investors demand higher yields at the long end of the curve when there
is economic growth and the prospect of future inflation. Or the yield curve
could also rise because short term interest rates are falling. Currently, neither
of these scenarios is occurring. This flattish yield curve suggests a slowing
economy where large capitalized growth stocks should outperform the more speculative
issues like the semiconductors.
Looking at figure #5, the top panel is a daily chart of the NASDAQ composite.
The bottom panel is 14 day rate of change indicator of the yield curve. The
yield curve is calculated by taking the yield on the 10 year treasury and dividing
it by the yield on the 3 month T- bill. Today's ratio is then compared to the
value 14 days ago. I have highlighted with blue ovals the six major "trade-able" bottoms
since 2000. Each was associated with a rise in the yield curve. The yield curve
has not confirmed this latest bounce. Is the rise in stock prices forecasting
better times ahead? The action in the bond market suggests otherwise.
Figure#5/ Yield Curve/ daily

Another thing that I am monitoring is the asset flows into the electronics
sector fund offered by Rydex. Once again, when these market timers generally
oversubscribe to an issue, the party is about to end. As seen in figure #6
below, absolute levels are not high but the relative levels are in an extreme
overbought position. I have highlighted other times through out the past year
where this was the case, and in general, these highs have led to some short
to intermediate term sell offs. The one instance that was not true to form
was the thrust off the bottom in March, 2003. So may be this time will be different.
Figure#6/ Rydex Electronic Sector Assets/ daily

Ok what would make more bullish? Three things: 1) a rising yield curve;
2) a mild pullback in the semiconductor index; 3) Rydex market timers
continuing to remain on the sidelines.
That is The Technical Take, and I hope you have found this analysis informative
and profitable!