Sunday, April 4, 2010

GGT Timers and Other Observations, Updated Easter Weekend

.
Summary

*** Remember -- these are snapshots in time looking backward, not a prediction of the future ***.  Some call this "driving by the rear-view mirror".
  • The GGT LCR Change Timer (fast) is LONG (+1).
  • The GGT LCR Change Timer (slow/confirming) is in transition at CASH-LONG (0)
  • The market is continuing to favor smaller caps relative to larger caps
  • The EMAs on the 65d slope of the VIX have all turned up, which is bearish
  • The number of "Longs" has fallen off the March peaks, indicating upside potential (fuel) but also showing that (if the past is any indicator of the future) that we could easily experience a significant pullback.
  • The slope of the EMA of the 65d price signal has just turned positive (65d EMA slope is now horiztonal) but has been moving upward the past few days from a negative value, which is bullish.
  • The EMAs on the LCR Change have all reversed and are now pointing upward (bullish), although they are still negative in value.  Watch for these to move into positive territory for a definite bull signal.
  • The GGT Price Index rates of change have broken a downward-pointing trendline are are now pointing upward, if not accelerating.  This is bullish.
  • Contrary to the talking heads, GGT Volume has been very stable and consistent with rising database prices.  This is bullish.
=================

GGT Long-Cash Ratio Change Timer Update

The GGT system uses two timers based on the Long-Cash Ratio.  These timers have three states:
  1. CASH (-1): You should be purchasing contra ETFs and have no long positions
  2. CASH-LONG or LONG-CASH (0):  You should be prepared for a state change.  CASH-LONG (0) means that you are previously in CASH and have a possible transition to LONG.  LONG-CASH (0) means that you are previously LONG and have a possible transition to CASH.
  3. LONG (+1):  You should be long in equities.
In my previous post I alerted you to a state change to CASH-LONG (0) of the fast timer which occured on Tuesday, March 30th.  I also alerted you to watch http://www.finviz.com/ to see if the ADV/DEC was supportive of such a move on Wed or Thurs, since I was going to be traveling both days.  Finally, I stated that the slow timer was still favoring CASH as of the market open Wednesday morning.

As of the close of Thursday's action we have the following:
  • GGT LCR fast timer has transitioned to LONG (+1).
  • GGT LCR slow timer has transitioned to CASH-LONG (0).
Until these two confirm we're in quasi-quasi-land on what to do.  Given this, sitting on cash with a slight bias to the upside is prudent, at least from the perspective of this timer pair.

Here's the equity curve of the fast timer, plotted with the GGT Price Index:



As with all of my graphs, you can right-click on the chart to open in another tab or in another window.

The left scale of the graph above refers to the GGT price index, which is shown in BLUE.  The purple trace is the GGT Equity Curve, using the fast timer, and it's scale is shown on the right.  Basically, if you would have put a $1 on the table and religously followed the GGT price index with buy and sell signals you would have doubled your money since January 2nd, 2009.  Note that this equity curve ignores the slower confirmation timer.

Two equations are shown in the graph, each colored corresponding to the curve they represent.  What is important to understand is that the coefficient of regression, known as R2 by some, is very, very good for the fast GGT dataset, giving a R2 value of 0.96 (on a scale of 1) for the data shown.  The way to interpret this is that the GGT equity curve is very linear with respect to time, which is what we want.

I've circled an area on the graph where the fast timer is struggling in the present market.  GGT was designed to correct itself over time, so my expectations are that it will continue upwards as soon as a definite market trend resumes.

The common complaint about the fast timer is that you can get whipsawed.  Very true, and thus the nature of a fast timer.  Hence I created the slower timer using the same model, which prevents a high rate of whipsaws at the expense of lower gains and higher drawdowns.  Here's the curve:




The first thing that should jump off the figure at you is that the gains of the equity curve are about half of the fast timer.  The second thing to realize is that this slower timer is also having difficulty in the present climate, hence the equity curve is falling a bit while the GGT price index is climbing.  Remember, the slower timer has just transitioned to CASH-LONG (0) from CASH, so it's not time to sell the contra positions (yet). 

Going forward from here ...

The rules of the GGT timer system say do the following:
  1. IF in a CASH (-1) situation, you should be playing the contra side of ETFs or shorting stocks.  Your longs are going to be beaten over the longer haul, so give consideration to closing them.
  2. IF in a CASH-LONG (0) situation, you have been in cash and are now in the process of transitioning to LONG (+1).  Wait until you have strong confirmation that the next trading day is going to be a very positive day according to ADV/DEC lines.  As I've instructed, the home page at http://www.finviz.com/ can provide you a very easy method of determining this (upper left corner of the home page).  IF the ADV/DEC ratio is at least 2:1 by mid day it *may* be safe to enter, e.g., GGT LCR Change will confirm a move to LONG that evening.  The problem in recent days is simply the number of reversals that we've been seeing -- this can all fall apart too.  A safer method is to wait until a firm LONG (+1) signal is issued with the close of the day.
  3. IF in a LONG (+1) situation, you should be playing the long side of equities.  Your contra positions are going to get beaten over the longer haul, so give consideration to closing them.
  4. IF in a LONG-CASH (0) situation, you have been LONG and are now in the process of transitioning to CASH (-1).  Wait until you have strong confirmation that the next trading day is going to be negative with respect to ADV/DEC lines, e.g., the ratio is better than 1:2 to the downside.
Monday's futures are up just slightly as I write this Sunday evening.  The FAST LCR Timer is LONG (+1) and the SLOW LCR Timer is CASH-LONG (0), which could move to the long side (+1) if Monday is an up day on equities.  Conversely, since Thursday was up according the the LCR, IF Monday is down according to the LCR the slower LCR timer will remain at CASH-LONG and the faster LCR timer could actually whipsaw back to LONG-CASH (0).  Stay tuned, and ask questions if you don't understand what've I've written.

=================

Micro Caps vs. Large Caps

As many of you know, I like to watch the ratio of micro cap stocks to large caps stocks.  I do this with ETFs, plotting the IWC (Russell MicroCap Index) against the IWB (Russell LargeCap Index).  HGSI makes this easy:


The chart above, specifically the top pane , shows the ratio of the two ETFs.  Plotted against the solid blue trace is a red, dotted trace which is the 13d EMA of the blue line.  What the data shows is that microcaps are out-performing large caps in the present market, and have been doing so since late November.  The result is that we should be focusing our efforts on smaller-cap stocks, not larger (although largecap stocks will continue to perform, just will do so behind microcaps).

=================

The VIX

The VIX is showing signs of life, which means volatility is increasing.  Increasing volatility generally means increasing pessimism.  Here's the chart:



Of particular interest to me is that in the 65d Slope EMA pane (white background), all the EMAs on the slope of the 65d EMA price are now in an up-trend.  This is a HUGE warning sign.  Here are the dates when this last occurred:
  • 1/22/10
  • 10/29/09
  • 10/1/09
  • 8/6/09
  • 6/15/09
  • 2/19/09
Certainly, these are not perfect indicators of a downward slide, but they have flashed their signals in the past near local tops.  Just be advised.

=================

GGT Relative Ratios

The following graph should be familiar:


This graph shows the relative ratios of the different rankings of stocks within the GGT database.  I've circled periods where stocks that were ranked "Long" have come off their peak.  You can see that we had a sustained period in March-April 2009 where we remained at lofty levels before the pullback in May-June 2009.  You can also see on the right hand side of the graph that we have pulled back a considerable amount from the peaks early in March, indicating that we could continue to remain at these levels if not higher. 

Don't miss one thing though:  EVERY time we hit a high, we had a significant pull-back within a month or two.  Make sure you understand the graph, specifically the yellow ("Long"), the blue ("Cash"), and purple ("Affirmed Cash") levels.

=================

GGT Price 65d EMA Change

The change in slope of the 65d EMA of the GGT Price gives us a good view of the intermediate trend for the database.  Here's the graph:



Plotted above is the GGT Price Index, along with the day-to-day change in the slope of the 65d EMA of the GGT Price Index.  Note that the change in slope is not smoothed -- this is "real time".

The first thing that should catch your eye is that starting about mid-March we started to decelerate in the day-for-day changes of the slope, which means that we had all indicators that a reversal was pending.  As soon as this indicator moved negative the slope of the 65d line on price moved from slightly upward to slightly downward pointing, another bearish sign.  Now, with the indicator reversing (which is bullish) and now transitioning to just above $0.00 (which too is bullish), the intermediate-term trend appears to be stabilizing (horizontal, no significant price appreciation/depreciation), which puts us in the quasi-quasi land I mentioned above.

Until we get some signficant movement above or below the $0.00 level caution is advised, although your bias should be to the long side.

=================

GGT Price and GGT Long-Cash Ratio Rates of Change

This next graph plots the GGT Price Index along with the LCR rates of change of various EMAs.  ROCs are interesting because as they transition, they can tell us about changes in market sentiment.



This graph reveals that the LCR ROCs were all negative, indicating that the bears were winning even though the database was moving to new highs.  This is a divergence -- either the price must reverse to sync with the increasing number of stocks moving to CASH, -- OR -- the LCR must reverse to sync with stock prices moving higher.

We have a perfect example of the latter occuring -- stock prices in the database continued higher, and the rate of change of stocks moving to a CASH recommendation slowed then reversed over the past week.  The fact that they are all still below $0.00 is bearish, but this is strongly negated by the fact that they are now pointing upward.  A cross-over of these ROCs into positive territory would be a strong bullish sign.  Hence this indicator is mildly bullish and gaining strength as we move onward in time.

=================
GGT Price and Price Rates of Change



The graphic above shows that over the last week that the prices in the GGT database resumed an upward move in appreciation, albeit a slight acceleration to the up side.  Breaking the downward-pointing trendline is bullish, and this graphic says that not only are we appreciating in prices (upward-pointing price movement), we are now doing so at a faster rate than over recent previous weeks.  Color this bullish.

=================

GGT Price and Volume

Contrary to the talking heads, I'm not seeing the "dry up" of volume others are seeing.  Remember that the GGT database is comprised of:
  1. Stocks over $1
  2. Stock only on the AmEx, NASDAQ, and NYSE
  3. Stocks with 50d MA > 50k shares
Perhaps the talking heads are looking at the universe of stocks, including OTC:BB and .PK stocks to draw their conclusions .... I simply don't know and more importantly, I don't care.






























The Graphic above shows that volume has been very steady since just before the January 2010 decline, which I consider healthy.  I'd be throwing red flags everywhere if we saw rising GGT prices on declining volume, but I'm not seeing it here.

Color the Price/Volume combination bullish.
=================

Remember, you are responsible for your own trading decisions, not me.  Please do your own homework.

Also remember that there is a general meeting of the NoVA VV group this Saturday.  Check the Yahoo! site for further information.

Regards,

pgd