NASA and other Astronomer-Photographers often use an image enhancement technique of stacking images taken at different times. Through this process they can increase the resolution beyond that of the original camera. A single photo may show a number of dim specks of light, but when stacked the dim stars are enhanced and "fake" stars that may be dust, gas or reflections tend to cancel out.  We too can, in a sense,  take photos of the stock market. When we do this, our forecasts appear similar to dim stars. We are now working with three such  "photos".  We use numerical techniques rather than graphic, but the results are similar.

1999 was a strong up year. Looking at the 1999 data [that was not used to train the program] we look at what happened when all three images were in agreement. (All graphs below use a 2x dynamic fund factor.)

The above graph shows a very consistent equity gain  when all three signals agreed.

The above graph shows what would have happened in 1999 with both two signals and three signals agreed.  The % correct was not as high here but the return doubled.  Now lets look at the recent time frame from Nov. 7, 2000 to March 16, 2001below.

We can see that the program is still working fine as long as all three signals agree. So let's see what happens when we combine two signals of agreement with the three signals, see below.

Under these extremely negative conditions two signals in agreement is not enough!  A quick look at the table below provides us with even more assurance.

It becomes very clear looking at both the % correct and the gain/loss that the reliability increases as we move from one signal to two signals and on to three. We have gone into this a lot further examining the up signals verses the down signals.

These numbers would be more even, but with 1999 being a year of excess up, and Nov-March 2001 reflecting excess down, the results skew. The skewing comes from what happens after the markets close. During a strong up period like 1999 the after-close reports are mostly good, and that turns weak buy signals into strong buy signals. Weak sell signals would become buys, but since the program was trained over mostly up markets weak sell signals are somewhat compensated for, see below.

The above figures show the average % per trade gained or lost. You can see by this that "buy" signals during bear markets are subject to the greatest risk. In extremely negative markets like this the weaker "buy" signals are subject to the overnight influences of negative reports from companies and brokerage houses. The answer...In extreme negative conditions trade in only the strongest signals until conditions improve.  Meanwhile we are busy building new "snapshots"