Debate on Astrology (AN v HWS)

[natureboy]

this is a good topic

@the hare - absolutely on the random. There are times where the astro correlations are present, and other times where (at least the techniques i'm aware of) they fall out of sync

@hwsteele - not knowing what we know - i tip my hat to you sir. From "Beginners Guide to Constructing the Universe" ~ "the power which we seek is the power with which we seek"... i got a little fired up, apologies to the last guy with Einstein.

@intuition - this is interesting, i'm looking into this blog.

Here are some times (EST) for EURUSD for this upcoming Monday 5/23:
5:40
10:40
12:40
14:50
these are from lunar cycle harmonic and natal data for the EC currency future. pivots often form within 20m or so of these times.

 

[enragedcow]

i think if astrology (god forbid) becomes ingrained in society then the movement of planets will have a self fulfilling effect on the markets

but our markets are most certainly not influenced by movements of planets/stars/black holes/moons/satellites/asteroid belts/dust/random crap up there or the higgs boson

astronomy is fascinating, astrology is lies, astrophysics is the key to becoming a god

 

[Splitlink]

i think if astrology (god forbid) becomes ingrained in society then the movement of planets will have a self fulfilling effect on the markets

but our markets are most certainly not influenced by movements of planets/stars/black holes/moons/satellites/asteroid belts/dust/random crap up there or the higgs boson

astronomy is fascinating, astrology is lies, astrophysics is the key to becoming a god

Absolutely right on all points.

Or would it be safer to say that I agree on all points?

 

[Vorbis]

Absolutely right on all points.

Or would it be safer to say that I agree on all points?

The Absolute, by definition, includes and transcends all variations of right and wrong. But I would agree that these points are 'relatively' right.

 

[Splitlink]

The Absolute, by definition, includes and transcends all variations of right and wrong. But I would agree that these points are 'relatively' right.

If you go into the Theory of Relativity I'll go into orbit and become "lunatico", myself! But I don't think that it will make me any richer.

 

[andy213]

with the greatest of respect, mathematics and science are the innovations and advances of the recent centuries, wheres witchcraft hocus pocus stars voodoo astrology and bone magic are relics of a tribal past now thankfully behind us. Modern statistics, ideas like binomial distirbution, probability etc completely disprove all perceived "links" between what time the sun came up one day and the price on a commodity the next! We can make these links until the cows come home. I have observed that every time I can see sirrus, it is night time, that must therefore mean that sirrus is a bringer of night!

Incidentally I made a few quid on a long last Friday 13th not so long go. Long? Friday 13th? surely not

TA however, is something I cannot assign fully to science or voodoo, I find it to be both. In it's purest form it is absolute abject horse s**t , but the clever twist is that people believe this to a point where their buying selling pressure allows voodoo to almost shoulder barge its way to a place in science. I find TA to work and to work well , because of this herd instinct and the fact it's a self fulfilling prophecy

so the kindest I am prepared to be to astrology is to say the same - if enough people place their longs or shorts based on where venus or whatever other celestial object is in the sky, then maybe, just maybe....

 

[natureboy]

Certainty? In trading?

why do people reject outright something they have not researched at all?

 

[natureboy]

@andy 213 - well, if TA only works b/c of "self-fulfilling prophecy", then how can you explain the same Fib ratio support/resistance relationships, and purely objective patterns on Cotton data from the late 1800's or stock indexes or stocks from the first quarter of the 20th century?

Before anybody knew! the behavior was already there...

 

[forker]

I don't mean to rain on anyones parade but I have to ask you a question. Do you see any institutional traders risking millions on astrology?

They are the ones that makeup most of the market and I am curious why anyone would attempt looking at the market in such an unusual way.

 

[the hare]

I don't mean to rain on anyones parade but I have to ask you a question. Do you see any institutional traders risking millions on astrology? .

IIRC a reasonably successful fund manager recently admitted that he selected a bias on market direction based on the color of the traffic lights he encountered on his drive in to work. If the lights where predominantly green, he was long, and if red, short.

I suspect that if institutional traders where risking client funds on astrology they'd probably keep quiet about it.

 

[natureboy]

@forker - JP Morgan used astrology. He could qualify as "an institution"

Its more about correlations that allow a trading edge, rather than "does it look ok" or "does this makes sense at a glance"

 

[j b]

@forker - JP Morgan used astrology. He could qualify as "an institution"

Its more about correlations that allow a trading edge, rather than "does it look ok" or "does this makes sense at a glance"

How do you go about testing the statistical significance of these correlations, in order to determine whether or not they are due to random chance?

 

[natureboy]

Law of Large Numbers and Relative Frequency, is how i approach it. Grid search.. The historical population in financial time series is limited, only going back so far, about 10 years for intraday EURUSD for example. These astrocycle times may give average of 7 time windows per day, thats 10 years*249 trade days per year*7 times per day = 17,430 events in the historical population in the case of EURUSD intraday data. So, set the confidence interval, figure the statistically significant sample size, randomly sample, and evaluate the tests!

The relative frequency if the astrocycles have no predictive power would be 1/2 or less of the observed time windows coincide with trend change as you define it.

If out of a large enough sample of historical instances the rule worked, then you can safely infer that the correlation is not due to chance - BUT ITS ONLY AN INFERENCE ---- you said "determine whether or not they ARE due to random chance" ---- you can't know that!!! call in any statistician, the best you'll be able to do is have a high probability that whatever rule/correlation is not just random chance, you can never eliminate the possibility entirely

 

[coopster]

Most people have no actual understanding of Most of what they "know".

Isn't this the point though? Markets are moved by the mass behaviour of people who as you say, may be influenced by sunlight and the moon etc.

But some are using a misunderstood concept of astrology, some may be using the "real " astrology. People will be trading the same market where one has day, the other has night and they both could be different distances from the moon. Maybe one trades in an office with no windows, the other from the deck of his yacht in the bahamas.

If there are more people trading the wrong astrology, then that will have more influence of the market than real astrology. I am pretty sure there are more people trading support and resistance than both of those put together so that will have more of an effect again.

My conclusion is - Do what works for you. I'm still looking but i know that i don't trust astrology due to woofters like Russel Grant telling me what my day will be like while using an out of date calendar as his reference.

 

[natureboy]

do what works for you - yes. like hwsteele said, people trade successfully with and without it.

different astro techniques give different results, no doubt. and like anything, there are charlatans, bunk methods, however do not affirm the consequent -> liars have a head, this person has a head, therefore this person is a liar? astro can work... so can entering on a coin toss. doesn't mean all astro is random, though ...its pretty wild, kind of mind blowing if its highly significantly nonrandom