Saturday, August 30, 2008

Trading The MACD

I wanted to add the MACD to the System Tracker earlier this year, but there seems to be so much variation in interpreting the signals that I needed to confirm some tests that I have been running before presenting it. Anyone who uses the MACD should read Gerald Appel's book, Technical Analysis. In it he gives his rules for the MACD. The only objection I have to some of his examples is that in several cases he is buying against the trend, the trend being the direction of the 50dema.

For myself, I have made a couple of simplifications which I believe work very well. Nevertheless, one should not regard it as a purely mechanical method. Some discretion is necessary, I think, in determining whether the chart pattern has the right look. This is hard for me to quantify since I have spent so much time studying elliott wave that oftentimes I have an intuitive sense that a particular signal doesn't look right.

Following are my rules for using the MACD:

1. The trend is determined by the direction of the 50dema.
2. The strength of the trend is determined by the angle of ascent or descent of the 50dema.
3. In general, when the trend is flat or weak to moderate use the 12-26-9 MACD for buying or shorting, as well as for exits.
4. When the trend is strong use the 8-25-6 MACD for buying or shorting and the 12-26-9 or the 19-39-9 MACD for exits.
5. In general, the MACD must fall below the zero line for a buy signal to be valid and above the zero line for a sell signal to be valid.
6. Buy positive divergences and sell/short negative divergences in the 12-26-9 MACD.
7. If the market is breaking out to new highs or rises from below to above the 50dema, then exit on the second sell signal - not the first. Reverse for shorts.
8. If the trend is flat, buy on the second buy signal from a higher low or with a positive divergence. If there is a lower low, wait for a pullback to confirm the signal. Reverse for shorts.
9. Set a stop loss at a minimum of 2 ATRs below the entry, but at least below the prior low.
10. Trail a stop loss at a minimum of 1 ATR below pullbacks to the 50dema.

It is difficult to find circumstances where rule 4 applies; however, two recent examples are POT and MOS, which had several signals in a row.

I believe that rules 7 and 8 are probably the most important for applying the MACD successully, which requires some chart reading experience. Oftentimes, the first sell signal is premature and even if the next sell signal is from a lower high, that high will frequently retest the prior high so that there is not much profit given up. The first buy signal in a weak or flat trend is often a false signal. Waiting for the second signal to confirm cuts down on whipsaws.

I will provide some chart examples in future posts.

Currently this year, the MACD is outperforming all of the other systems in the System Tracker with an amazing 34.21% gain with only 3 trades. The ARR is 51.32%. The total time in the market has only been about 15 weeks out of the 34 weeks year to date.

At the present time the Qs are in a weak uptrend. The last buy signal occurred with the 50dema in a downtrend so it was only a short exit signal not a new long entry signal. The last sell signal occurred with a weakly rising 50dema, so it was not a valid short entry signal. Since the next buy signal is likely to find the 50dema flat, it will not be a valid buy signal. It will take at least one and perhaps two more signal before there is a valid entry point. If the next sell signal is from a lower high, then that would be a valid short signal. This may take one to two more weeks.

If you are currently trading actively, but achieving lower returns, you may want to consider using the MACD with the above rules and simplify your trading style.

Monday, August 25, 2008

Still A Pullback

While the markets closed down significantly today, volume overall was light. The Qs have closed below the 30 and 50 demas after a 6 day pullback. Look back at April and you will see the same setup. I do not expect to see a rally equivalent to the second half of April and May, but rather something more like the end of December when a 5 day rally ended the countertrend rally from the November low.

Be ready to exit long positions on strength in the next few days. If the markets do breakout that's ok. It would be a great place to exit. You can always re-enter on a successful test of the breakout support.

Thursday, August 21, 2008

Buying Opportunity

This week's pullback in the Nasdaq 100 QQQQ and the Russell 2000 IWM is a buying opportunity. The 15 dema crossed up the 30 dema on 8/11/08 and these markets are now pulling back to support at the 30 dema. The Qs are also sitting at support on the 200 dema and the 50 dema while the 50 dema is about to cross up the 200 dema. A simple entry trigger would be to buy above the last two days high, currently at 47.59 in the Qs and 73.83 in the IWM.

Sentiment as measured by various gauges has not yet reached an overbought extreme, and the negative drumbeat on the financials continues to keep the pessimistic mood alive. So, not only do bulls have sentiment on their side from a contrarian point of view, but the upswing in the 22 week cycle (second half of the 10 month cycle) has not yet peaked.

The lagging performance of the SP-500 and the Dow 30 cautions that once this rally is over lowers prices will be seen.

Targets are 51.47 for the Qs and 79.84 for the IWM.

As a side note the volatile VLE system has not yet given a sell signal with this pullback.

Tuesday, August 19, 2008

Was Friday The Top Of This Rally?

Probably not. Monday's decline was not a distribution day as volume declined on the major indexes. NYSE new lows was only 39, less than the 40 level that could indicate trouble. Support for the Qs is between 46.50 and 47.00 at the rising 50 and 200 demas.

This week may be choppy to the downside, but unless support is broken at 46.15, we should see higher prices before this rally is over. Look for a move to new rally highs beginning sometime late this week or early next week. Negative momentum divergences and breadth deteriotion at that time will confirm that a top is at hand.

Monday, August 18, 2008

Pullback Of Acceleration?

The Nasdaq 100 (QQQQ) and the Russell 2000 (IWM) did not continue the pullback into Friday's expected turn date, but rather continued to new rally highs ending the week with a narrow range bar (smallest range for the last 7 days) for the Qs and a double top reversal from the June high for the IWM. Both of the markets appear setup for a pullback to the 50ema.

However, other major markets remained below the rally highs into Friday. The mixed condition makes it unclear if we will see a pullback or rather if the narrow range bar will lead to an acceleration breakout.

The best course of action is to wait and see how the markets develop this morning before making any trading decisions.

Gold has possibly found support at a two year trendline and may be ready to bounce for a few days. New lows or a test of Friday's low is likely.

Wednesday, August 13, 2008

Pullback To Expected Low

As was mentioned in an earlier post, August 16 is an important low from last year. At this point it appears that the markets are consolidating the recent advance into that time period, which is bullish. The QQQQ 15ema has crossed up the 30ema, however, a little more consolidation or pullback would be nice before trying to go long (if not already long). The measured move (March low to June high) target from the July 15 low is 52.86. The Jan 08 high is 51.47. So it would be reasonable to expect a high around that price band. In elliott wave terms, such a high would represent intermediate wave (2) or (B) of the decline from the October 07 high. This means the meat of the decline is yet ahead of us, so don't overstay your welcome in this rally. The most likely time period for a high is around 9/4 to 9/9/08.

Using the same measure move method, we can now project the next low to be around 38.84 to 30.18 in the QQQQ unless the October 07 high is exceeded.

Sunday, August 10, 2008

Technical Rally Underway

Have you heard that phrase before? What does "technical rally" mean anyway? In this case, it means that the rally is occurring on below average volume and less than desireable breadth while price level buy signals have triggered. Even Friday's large rally was accompanied by 56 NYSE New Lows and NYSE New Lows has only been below 40 3 times since this rally began on July 15. You may remember that the Cabot's Two Second Indicator looks for New Lows to be below 40 for healthy rallies. Volume was also well below average on Friday and less than the volume on Thursday's selloff.

This doesn't mean that traders can't profit from the rally, but rather that one should have a clear understanding that unless a significant volume shows up by the end of August that the markets will likely retest and move below the July lows. Traders should be ready to exit positions on any sign that the rally is losing strength and expect whipsaws like Thursday and Friday to continue to be the norm.

So what clues can we look for to let us know that the rally is ending?

1. McClellan Summation Index turns down.
2. % stocks above 40ma exceeds 60 and turns down.
3. Absolute Breadth reaches an extreme level - below 20 or greater than 55.
4. VIX & QQV breakout above the upper trendline.
5. Consecutive weekly closes in the lower half of the weekly range.
6. Failure to follow-through at key moving averages - 50dema and 200dema.
7. Negative divergences in the RSI and MACD.
8. Distribution Days increase to 5 or more.
9. Leading Stocks of the rally fail.
10. The financials start breaking down.

Keep a list of these types of indicators and look of the multiple confirmation that the rally is ending.

Wednesday, August 6, 2008

Buy Signal?

Today the Modified Donchian system confirmed a buy signal in the Qs, and while the Weekly-Daily system signaled an exit from the existing short position, today's volume did not confirm a buy signal. The Cabot's Tides also signaled an exit from the existing short position.

Above is a weekly chart of the Qs with the 1.0 SD and 2.0 SD 20 period Bollinger Bands. This rally has the feel and look of the December rally in the Qs. Notice the top that occured at the 1.0 band in early December. If this rally continues to meander up toward a high around the 15th on diminishing range and volume, look for a top around the 1.0 band followed by a sharp decline which may be followed by another sharp rally similar to the one in December.

While the weekly macd is turning up, we are possibly several weeks away from confirming a new bull market as opposed to a continuation of the current bear market.

Tuesday, August 5, 2008

New Buy Signal At Hand

A move tomorrow above 46.15 in the Qs on above average volume will trigger a buy signal using the Weekly- Daily system as the Qs will move above the high of the low month (July). If the Qs close above 46.15, it will trigger a buy signal using the Modified Donchian system. A first target for the Qs is 47.82, which is the .618 retracement from the July low. If that level is exceeded on good volume, then we may expect that the June high will be tested if not taken out.

Unless volume expands significantly before 47.82 is reached, I would look to take profits at that level. You can always re-enter later if the market moves higher.

Sunday, August 3, 2008

Rally Attempt Underway

(Click to enlarge)

It's been awhile since my last post. I must say that I admire all the bloggers that can find the time to post daily reports. Lately, I have been pushed with the demands of business and family commitments. However, we are now at an important juncture and I want to share my thoughts.

IBD confirmed a market rally on 7/29/08. (The System Tracker has been updated.) Even though the Qs have not pushed through resistance at 46.15, it certainly would not be prudent to enter new short positions here. Some analysts are calling for an immediate return to the bear market based on the weakness of the rally, while others are waving the All-Clear sign, but it's too early to say one way or the other.

One fact that I would like to bring to your attention is how often markets turn near anniversary dates of previous highs and lows. The Qs topped on 7/19/07. We just had a bottom on 7/15/08 4 days before the anniversary date. I have seen this phenomenon too many times to dismiss it as coincidence, so I look for confluences of projected cycle highs and lows with these dates to help pinpoint tops and bottoms.

I am not suggesting that you should try to trade directly off of these dates, but look for potential reversal signals around these confluences. When market sentiment is extreme, it may be reasonable to buy a bottom as was the case recently with the financials. The UYG is now up 56.6% from its low. With the bad news around Fannie Mae and Freddie Mac on 7/11/08, I took a position in the UYG with an expectation of a similar reaction after the Bear Stearns event. I added to that position on 7/15/08. My target is the 26 to 30 range. I don't typically take this type of trade, but in this case there were just too many factors suggesting a bottom was at hand to pass it up.

But what can we expect from this point forward. The market bottomed last year on 8/16/07, so we could see a top or bottom around that date. Currently, we are in the upswing of the 22 week cycle with a top expected around September 9. However, given the lateness of the previous 22 week cycle low, I would not count on the market holding up to that date. If the Qs continue up to 8/16, I would expect a quick pullback before a final top around 9/4 to 9/9. If the Qs move down into 8/16, then I would expect a sharp rally into the top.

In either case, don't count on too much upside. If the May/June highs are not taken out in September, look for a nasty selloff this fall. Look at the trend following systems in the System Tracker for a guide to entries and exits around these dates. Most of the systems are still short, however, most are near exit and/or long signals.

Good trading.