2011 10.30 Weekly DJIA Long $ Short Term

After updating, the model discovered a short period cycle resulting in a much better overall fit. Had the model not converged, it would have been thrown out.  There is still a high discrepancy between the actual and predicted price, though not quite large as before.

Visit econocasts.net for free sample trading model downloads. 

2011.10.28 Interim Gold Long $ Short Term

From this model's perspective, this is an almost six-sigma event.

This is so weird that I'm going to hold off judging whether the model is valid.  A week or two and the answer should be obvious. Maybe.

Visit econocasts.net for free sample trading model downloads. 

2011.10.26 Interim Gold $ Silver Short Term

Both of these models have reached extreme values of disparity between the actual price and the predicted price. Based on information from a completely different model for SLV I use to guide investment stance,  my bet would be that silver will be the one that breaks down first but GLD is still in "buy mode" as of this moment.The concordance of the two models from the two time series supports validity, or the same structural flaw is present in the two models. This is unlikely but....

So, the possibilities remain:

• the model is valid (look out below! soon!)
• we are experiencing a "six-sigma" event but the model is valid (look out below, at some point)
• the model is not valid 

Visit econocasts.net for free sample trading model downloads. 

2011.10.21 Weekly Silver Long $ Short Term

Visit econocasts.net for free sample trading model downloads.

2011.10.15 DJIA 1928-1932 and 2004-2012

click on images to enlarge

Rather than another tedious 1929 to the present comparison, here is a less tedious one. I wanted to highlight a couple of points. The first is that the spread between the actual price and predicted price, as measured by the yellow Z-score has now broken the record for historical extremes on the upside.

The second point is made by the red circles on the charts, showing where flat patterns developed during relief rallies just after 1929.  The technical term for this is "frog jumping down steep hill" also known as "the toad tilt."

These time series are multifractal, as discovered by Benoit Mandelbrot and others.  This means that scaling may occur simultaneously along the time and the price axis. In other words, one pattern may match another by stretching or compressing both time and price scales.

If this model is correct, we should soon be traveling down a steep slope, probably by next week. If prices increase much more, or even stay relatively static for a few trading days, and the Z-score goes to >5, then I'll have to retire this structural equation cycle model.  Either way, there is a lot to be learned depending on what happens next week.

I'll be answering some thoughtful posts over the weekend, good luck and good health to all!

Visit econocasts.net for free sample trading model downloads. 

2011.10.14 DJIA: Long $ Short Term Predictions

click on images to enlarge

In honor of this historic occasion, I have represented the very long term DJIA chart above.

If I was long this market, I would be very very afraid.  According to this model, this level of divergence was last seen in 1929 where the actual price is so much higher than the predicted price. There are other instances of higher divergences, but they have always occurred on the downside, not on the upside.  Of course, "this time it may be different."

I would rate the chance of a significant correction next week to be extremely high. Of course, it is also possible that this particular cycle model may be ready for /dev/null or the bitbucket, or the dustbin.  I don't think so - but I'm biased.

Visit econocasts.net for free sample trading model downloads. 

2011.10.14 Gold: Long $ Short Term Predictions

click on images to enlarge

This is the fourth time this year that the divergence between the actual price and predicted price has reached z>2. All of those times were associated with reversals. The probability of a reversal is very high in the next 5 days +/-.

Visit econocasts.net for free sample trading model downloads. 

2011.10.13 Silver: Long $ Short Term Predictions

The last few times that the spread between the the actual price curve and the model prediction curve reached 2 SD units, a correction occurred in less than five trading days.

Links to previous silver cycle model charts

2011.10.07
2011.09.30
2011.09.23
2011.07.15
2011.06.26

Visit econocasts.net for free sample trading model downloads. 

2011.10.12 DJIA Z-score of cycle model differences

Click on the image to enlarge.  To keep the chart simple on the cycle models, I use the absolute value of the z-score, since it is a symmetrical function. In this case, the sign value (+ or -) shows the direction of the expected reversal. Thanks to Greenface over at daneric's site for asking the question that inspired this.

Viewed in a historical context, this EOD chart suggests we are approaching historical extremes not seen since the mid-50's as well as other economically unpleasant periods in the past.

Edit on 2011.110.12 at 16:30:  I prettied up the chart and used EOD DJIA data rather than the 2:00 PM market price level.  The story is the same.

Visit econocasts.net for free sample trading model downloads. 

Cycle Models: DJIA Long $ Short Term Predictions

Today, the DJIA actual price curve has moved almost 2 standard deviation units from the model curve, so here it is for the record.  This is an extreme value suggesting that a correction, which would bring the two curves together, is probably not far off.  Or this model is broken.  I'd vote for the former, but I'm biased.

Cycle Models Silver Long $ Short Term Predictions

No, this Silver model is not a mistake.  It is the latest structural equation model.  I have been doing silver model updates mid-week because I was expecting the possibility of a model bifurcation given the differences in long-term predictions between two usually correlated metals comparing the older silver model with the present gold model. I was rewarded early this morning having watched the "parameter screen" with utter amazement, feeling like I was watching a screen in the Matrix. The new model slowly collapsed two longer fractal cycles into one cycle.  This is always a Good Thing because the fewer parameters, the better the model.

If you look closely, on the short term side, there is no significant difference between the old and new silver model in the short-term.  Both predict a downdraft.   The new model shows a much different longer term prediction, which is more in line with the gold model, and in line with a deflationary depression. It was hard trying to see how gold could go one way and silver another for a prolonged period of time.  It turns out that in statistics speak, there is not much difference between the old and new silver model - we are simply looking at two possible outcomes of almost even probability.

Taken together, my observations on the two models point to the new silver model as having better predictive accuracy, but not because of model statistics - this is my opinion. I will continue to run both old and new silver models to see if the next few days the new model declares itself as statistically better.  Or not.  Meanwhile, I am posting the new silver cycle model as the default model.