Astro-Harmonics
Greg Meadors (a trader who at one point had a fairly good record in a trading contest) has posted the following to newsgroups as a good introduction to astro-harmonics:John Nelson was a radio propagation specialist for the R.C.A. Company in the 1940's and '50's. His job was to predict radio disruptions from sunspots, magnetic storms, etc. Using normal cyclical analyses, he was able to achieve an accuracy of 65%. Trying everything, he could not get any better.I picked up a copy of this book (via Amazon books), a slim, old-fashioned paperback published by the American Federation of Astrologers. Here are the dates of the 13 worst storms referred to, as well as an additional date mentioned by Nelson as the greatest cosmic ray shower in history:Someone suggested that he look at planetary positions relative to the SUN, i.e. Heliocentric Astrology. He was an engineer and had NO interest in astrology. He decided to at least look at it, as there MAY be some direct effect. He took the 13 worst Magnetic Storms on record with RCA and drew the planetary positions at the time of onset.
He was surprised and shocked to find that many of these planetary positions were in harmonic angles to each other at the onset of disturbances.
Harmonic One = 2 planets conjunct (same space)
Harmonic Two = 2 planets opposite in the sky
Harmonic Three= 2 planets One Third around the sky from each other
Harmonic Four = 2 planets Square (fourth part of the circle)
Studying these relationships over time, he was able to predict radio propagation quality at 6-hour intervals a month in advance with 88% accuracy! He wrote a book on it called "COSMIC PATTERNS".
I took the dates of the 13 worst storms from Nelson's book and ran statistics on % change in the DJIA from -10 days to +10 days from the onset of magnetic storms. The stock market has a meaningful propensity to DECLINE from ONSET -2 to ONSET +3 days.
Nelson used to call me when a storm was in effect, confirmed by his Short Wave Set disturbance. I would then call a broker and find that very often stock prices would be dropping and Gold prices rising! Using Heliocentric Alignments, I have been able to predict, in the popular press, and weeks in advance, the dates of sharp single day market declines.
We both found that, if a HIGH TIDAL FORCE (New Moon or Full Moon) was in effect, the results in both Short Waves AND Stock Prices would be sharply amplified, as the Ionized upper layers of the atmosphere (Ionosphere) were brought much closer to the Earth by the Tidal Effects!
03-23-40 08-04-41 09-18-41 10-02-42 02-07-44 03-27-45 09-23-57 04-24-60 07-15-60 08-30-60 11-12-60 04-14-61 09-22-63 02-23-56 (greatest cosmic ray shower in history)One thing I would mention first off is that Meadors refers to Nelson's work as astrology, when I think that it might be better referred to as astronomy, simply because Nelson's application of the data does not necessarily involve the idea of supernatural forces.
The basic idea behind Nelson's work is that sunspot activity (which does have significant effects on the amount of ultraviolet radiation reaching the earth) is affected by certain planetary positions. I am not sure of the validity of Nelson's claims, but I thought it would be fun to take a look at it.
The assumption is that the gravity of the planets distorts the sun in such a way as to affect the formation of sunspots. I'm no physicist, but who knows, maybe the sun, being somewhat fluid, could be distorted in this way, even though the total effects of the planets would be rather inconsequential in comparison with the sun's own gravity. But maybe it's enough to put it over the edge. One thing I did was to calculate the gravitational forces between the sun and each of the planets. Here I have ranked the planets by their gravitational effect, in terms of the proportion of the total gravitational effects of all planets.
Jupiter 74% Venus 10% Saturn 7% Earth 6% Mercury 2% (note: the total of the rest of the planets combined amounts to less than one percent)Actually, after calculating this, I find it interesting that sunspot activity has a cycle of roughly 11 years, and that Jupiter (the planet that should clearly have the greatest affect), circles the sun every 11.86 years. Of course, as a skeptic, I should remind myself that this could be a coincidence. After all, it would make more sense if people had identified a roughly 12 year sunspot cycle. A 22 year cycle is also mentioned for sunspots, and we would expect this to be nearly 24 years based on the Jupiter theory.
In terms of the effects of the planets, I notice that "astro-harmonicists" sometimes bring Pluto into their calculations. Pluto's gravity is almost totally inconsequential. Remember that Pluto is not only smaller than our moon, but is about 30 times further away from the sun. In addition, there are other satellites of equal effect as Pluto, but they are not classified as planets. Why aren't those mentioned? Arch Crawford (another "astro-harmonicist" with a good track record) defended Pluto on the grounds that it must have a very strong electromagnetic force. Again, I'm no physicist, but I'm not sure we could get much of a force even if it was a chunk of solid magnetized iron.
OK. But what about effects on the stock market? I could possibly "buy" the sunspot / ionosphere argument, but what about the stock market? What did Meadors see when he looked at the data?
Well, taking the 13 worst storms listed above, as well as the "greatest cosmic ray shower in history", and plotting the stockmarket (Dow Jones) from 10 days before through 10 days after (adjusted for volatility), we get something like this:
The vertical axis in the center marks the day of the storm, and the heavy black line is the average of all 14 series. There does seem to be a propensity to decline, although I see it more as from -4 to +5 days, not the -2 to +3 days that Meadors refers to. However, let's stick to Meadors view.
Using 1000 resamplings on the 10 days before through 10 days after each of Meadors' 13 dates, I get a "confidence" of 88.7% for the decline over the -2 to +3 days, which is not bad, but not great by any means. Basically that means you could expect the same results 11.3% of the time by chance. At this level of confidence, I am mildly intrigued, but by no means convinced. If I extend the data to the entire period of 1940 through 1964 the confidence gets boosted to 95%.
However, as a skeptic, I'd like to point out that a 95% (2 sigma) confidence level means that we can achieve these results one out of twenty times just by random chance (note that for each harmonic there are 36 possible two-planet pairings of planets to be considered). By way of comparison, the superbowl indicator (the surprisingly strong correlation between the AFL vs. NFL winner of the superbowl and the the direction of the stock market for the next year) has a 99.7% confidence level. We would expect these type of results only 0.3% of the time by chance. However, I think it is fairly intuitive that it is by chance. In fact, what we really need to look at is how many millions of different events that people look at, and realize that something is bound to correlate.
Back to the astro data, one thing I notice is that the majority of the data, about 9 points, comes from bear market years (i.e. 1940, 1941, 1942, 1957, 1960) when there was already a bias to the downside in the market.
Also, as I look at the data, I see two bottoms within a day of the central date, as well as perhaps three tops, as well as many lines declining pretty much in the middle. Actually, I find the tops and bottoms more interesting than the declines, although I'm not sure that a possible inclination towards a turning point either way maybe one-third of the time is that compelling. But it is really too small of a sample to draw any conclusions, at any rate.
Actually, what I really wonder in regards to astro phenomenon, is why the US stock market? Why not the effects on trading in gold, the Japanese yen, wheat, or the Hang Seng stock market? What about used cars or new computer sales? What would single out the US stock market for this special treatment by ultraviolet radiation? Other than wishful thinking?
Because of some of my other pages, I have had some correspondence with a several astro proponents. Invariably, as I ask for "evidence" they tend to waffle and avoid answering the question. Two of them recommended I buy some astro software (costing in excess of $1000) for which they apparently would receive a sales commission. (Admittedly, one guy said he was going to give me a discount, perhaps that meant he was going to sell it sans commission. I don't know). At any rate, I'm just not prepared to throw away $1000 on a lark. Currently, the evidence presented above is the best I've come across, and it is a small sample with less than ideal confidence.
What I would be looking for, proof-wise, is a definable astro phenomenon that has an effect on the stock market. Let's pick some kind of configuration that happens often enough that we could get at least 100 events over the last say 50 years (of course, if there really were something extraordinarily profitable going on, would anybody be saying anything about it? - I think not). Just give me the dates, and I'll crunch the data. Preferably, I would need a way of generating the dates independently, to be sure you weren't just pulling my leg.
There is some C code floating out there for generating ephemeris data - if anyone is so inclined to put together an executable that would generate an ascii file of date and planet info, let me know, regardless of whether you are a believer or not. Then I could put it in a spreadsheet and crunch away.
I guess if nothing else, the astro approach is not correlated to other approaches - i.e. they're probably not doing what everybody else is doing.
This page is referenced in Van Tharp's book, Trade Your Way To Financial Freedom, chapter 5.