why have you accepted the price-predictability assumption? | Page 101 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

So I accepted a Prop Trading job in South Beach Miami 44 replies

Why you shouldn't show how much you have made in FX 35 replies

Incorporating Slippage Assumption in backtesting 4 replies

Market Predictability - Technical Signals 25 replies

Contesting the Accepted View of Bernanke 0 replies

Attachments: why have you accepted the price-predictability assumption?
Exit Attachments

why have you accepted the price-predictability assumption?

  • #2,001
  • Jan 20, 2014 3:38pm Jan 20, 2014 3:38pm
Quoting diceman555
{quote} hi pbylina, don't mention the trend word on this thread that's worth a good 100 pages of a free for all, killing field.
Lol, Ok.;-)
  • #2,002
  • Jan 21, 2014 12:07am Jan 21, 2014 12:07am
ooooooooooooo its all quiet, I just stuck my g/j trade in Simple Strategy room as I mentioned, and yes I made a mistake
SEE LINE,TRADE LINE..PRICE HAS TO GO SOMEWHERE,,, PRICE WILL GO SOMEWHERE.
  • #2,003
  • Jan 21, 2014 12:56am Jan 21, 2014 12:56am
I Just won 2 trades with at trade explorer using my automatized gambling system
"There's a sucker born every minute" - P.T. Barnum
  • #2,004
  • Jan 21, 2014 1:07am Jan 21, 2014 1:07am
Quoting Proximus
I Just won 2 trades with at trade explorer using my automatized gambling system
wayhay,, nice one bud
SEE LINE,TRADE LINE..PRICE HAS TO GO SOMEWHERE,,, PRICE WILL GO SOMEWHERE.
  • #2,005
  • Jan 21, 2014 1:21am Jan 21, 2014 1:21am
Quoting Proximus
{quote} Thanks, its 3 wins now, soon the 4th will come. Ok i`ll wait for your advice then.
one method I traded was a 2/20 ma cross on hour chart,actually you can forget the 2 ma, it is similar to your method but SLOWER, I was asked to do a trade earlier with g/j and I actually was trading an inside bar but I also used the 2/20 cross as well, managed to get 50 pips, it all fits
SEE LINE,TRADE LINE..PRICE HAS TO GO SOMEWHERE,,, PRICE WILL GO SOMEWHERE.
  • #2,006
  • Edited 2:42pm Jan 21, 2014 3:49am | Edited 2:42pm
eschewing CHF ( banks intervenes too much?)
NZD (too derivitive?).
Currency list:

Eur
usd
gbp
Aud
Jpy
cad
yields 5+4+3+2+1=15 pairs, all supported by my broker

1.EURUsd.
2.euraud.
3.eurcad.
4.eurgbp.
5 EURjpy.
6.USDcad.
7.audusd.
8.gbpusd.
9.usdjpy.

10.gbpcad.
11.gbpjpy.
12.gbpaud.
13.audcad.
14.audjpy
15.cadjpy

Each currency may be coupled with four others, therefore:
each pair has four parents that can push it around.

FIRST GROUP:

1.-EURUSD:
-2.euraud.& -7.audusd.
-3.eURcad&6.usdcad
-4.eurgbp& -8.gbpusd
-5.eurjpy& 9.usdjpy


2.EURAUD:
1,eurusd & - 7.audusd
3.eurcad& -13.audcad
4.eurgbp & 12.gbpaud
5 EURjpy.& - 14.audjpy



3.eurcad:
1.EURUsd&6.USDcad
2.euraud& 13.audcad
4.eurgbp.& 10.gbpcad.
5 EURjpy.& -15.cadjpy

-4.eurgbp:
-1.EURUsd.&8.gbpusd.
-2.euraud.&12.gbpaud.
-3.eurcad.&10.gbpcad
-5 EURjpy &11.gbpjpy.

.5 eurJpy
4.eurgbp &11.gbpjpy.
2.euraud &14.audjpy
3.eurcad& 15.cadjpy
1.EURUsd & 9.usdjpy

sECOND GROUP:

-6.USDcad:
7.audusd.& -13.audcad
8.gbpusd.& -10.gbpcad.
9.usdjpy & 15.cadjpy
1eurusd &-3.eurcad

7.audusd:
-6.USDcad & 13.audcad
8.gbpusd & -12.gbpaud
-9.usdjpy.& 14.audjpy
-2.euraud.& 1.EURUsd.

8.gbpusd:
-6.USDcad.& 10.gbpcad.
7.audusd.& 12.gbpaud.
-9.usdjpy. &11.gbpjpy.
-4.eurgbp.& 1.EURUsd.

9.usdjpy:
6.USDcad & 15.cadjpy
-7.audusd.& 14.audjpy
-8.gbpusd& 11.gbpjpy.
5 EURjpy.& -1.EURUsd.

THIRD GROUP
10.gbpcad:
11.gbpjpy & -15.cadjpy
12.gbpaud.&13.audcad.
6.USDcad & 8.gbpusd.
-4.eurgbp.& 3.eurcad.

-11. gbpjpy.
-10.gbpcad.& -15.cadjpy
-12.gbpaud.& -14.audjPY
4gbpjpy, & -eurjpy
-8.gbpusd. & -9.usdjpy.

12.gbpaud.
-13.audcad. & 10.gbpcad.
-14.audjpy & 11.gbpjpy
-4.eurgbp& 2 euraud
8.gbpusd. &- 7.audusd.

FOURTH GROUP:
-13.audcad:
-14.audjpy & 15.cadjpy
-6.USDcad. &7.audusd.
2.euraud.&3.eurcad
10.gbpcad.& 12.gbpaud.

-14.audjpy:
-13.audcad.& 15.cadjpy
2.euraud. & -5 EURjpy.
-7.audusd. & -9.usdjpy.
12.gbpaud. & -11.gbpjpy.

15cadjpy:
13.audcad. & -14.audjpy
3.eurcad. & -5 EURjpy.
6.USDcad. & -9.usdjpy.
10.gbpcad.& -11.gbpjpy.


(I have signed them the way I intend to trade them, according to ideas of hedging correlated pairs).
There are thus sixty separate known channels of influence of pairs upon one another in pairs offered by my broker that I watch.


It's complicated, but amongst the noise , One can see and trade on strong emergent signals that are not simple random chance.
That's 15 separate screens to watch with four each parent combinations effecting them.


I have made a chart that simplifies things somewhat, but busy, busy, busy.
Around news surprises, frantic, frantic frantic.


I guess I am going to be forced to Program an EA.
Attached File(s)
File Type: txt the forex garden bed.txt   2 KB | 379 downloads
  • #2,007
  • Jan 21, 2014 4:10am Jan 21, 2014 4:10am
The parent child numbering is indeed left some more scope of discussion.

In my view when I see, AUD/USD is in strong move, it is going to direct the other aud pairs and its crosses very soon. And we can place our trades likewise. With this we will have more good entries.

Attached Image (click to enlarge)
Click to Enlarge Name: Screenshot1.png Size: 116 KB
  • #2,008
  • Jan 21, 2014 4:32am Jan 21, 2014 4:32am
Quoting the redlion
I am curious on what justification do any of us have to have accepted the assumption of price predictability and market innefficiency? Is it faith or hope that in fact innefficiencies exist in the market and that we are actually able to profit? or do we actually have evidence and sufficient reason to support the asumption of an inneficient forex market? In economics and finance it is often said that profit comes from information, a well informed investor is able to profit from temporary market inneficiencies or mispricing of certain assets, if in...
As you can see from your post, efficient market relies on multiple participants who find inefficiencies and exploit them. Unless all inefficiencies have been found (which is quite a bold claim) there are still inefficiencies that can be exploited. New inefficiencies will rise if someone exploits old inefficiencies too heavily and this can be easily happen with multiple (and unknown number of) participants.

I personally believe there will always be inefficiencies that can be exploited but finding them comes much much more complicated when almost all the easiest are found.
  • #2,009
  • Edited 2:44pm Jan 21, 2014 5:23am | Edited 2:44pm
Quoting rekon67
The parent child numbering is indeed left some more scope of discussion. In my view when I see, AUD/USD is in strong move, it is going to direct the other aud pairs and its crosses very soon. And we can place our trades likewise. With this we will have more good entries. {image}

The numbering scheme In my last post, is quite arbitrary: The numbers are those of the windows I have open in Metatrader ( left to right, top to bottom) and little more than chance.

I am not at all sure that all the parent child relationships are represented without error. (later) In fact, there are several errors which I have correct in the post, but not in the file. reader beware!

I noticed when I was doing this that they seemed to arrange themselves in groups: One , I think around the EURO, One around the USD, one around the GBP, and one around the JPY (or is it the AUD?). There may be some significance in this, My instincts tell me I should investigate that.

There is weak , if any,current push from the parents on the AUDUSD: If it goes somewhere (it could well do that on its own), it will indeed effect may crosses. News from china (a currency not on my charts) can push this, so that it is more frequently an initiator of sequences rather than a child of them.

Good money can be made from this pair.

I am curious to see if it does get moving, if the parents will swing into more solid support for it. That would imply a regenerative feedback loop of which I have not yet been made aware...

I put so many screens up to make myself more aware of and open to good trade setups. this may be too many for others. They may wish to just watch one of the groups.

I left out the CHF and NZD currencies. How may parents could each child have if I included those? How many opportunities am I missing because of that, OR because of the Chinese Remini, the Indian Rupee, --- Shekels , anyone?

I am pointing out that this is a work in progress, which may never be completed, and could well get abandoned, should I decide that it is all a crock.
  • #2,010
  • Jan 21, 2014 2:17pm Jan 21, 2014 2:17pm
hi Rap, hope you are ok, good work you posted
SEE LINE,TRADE LINE..PRICE HAS TO GO SOMEWHERE,,, PRICE WILL GO SOMEWHERE.
  • #2,011
  • Jan 21, 2014 2:33pm Jan 21, 2014 2:33pm
Quoting diceman555
this research certainly touches on my own findings regarding correlated pairs and combined volatility.unable to find the full paper and also dated somewhat. Physicist Guannan Zhao, Ph.D. student at the University of Miami, and his collaborators have developed a mathematical model to describe the timing of price changes of currencies and the overall dynamics of the Foreign Exchange (FX) market. Zhao presented his findings in a keynote address on Saturday, March 12, at the second International Conference on Financial Theory and Engineering (ICFTE...

Thank you for sharing this paper. I believe they may possibly be confirming the heavily interlocked nature of pairs with common currencies.

I am tempted to build a calculator that takes the major parent currencies, measures and adds their changes, weighted with their respective sizes, and see how close it comes to the child pairs.
I suspect , closer than many would believe. How's that for price predictability. I have noted how , in general, the stronger the parents push the child pair, the faster and more dependably it moves.

Above a certain threshold, it seems to be virtually certain.

I am limited in my energies to do things like this. (74 yr old, surviving cancer)

It may well be that not only can future prices be predicted, they can frequently to a great degree of accuracy be calculated!

There . that describes my swing from almost all the way on one side, to the other on this question.

Waiting for a swing back...
  • #2,012
  • Jan 22, 2014 2:50am Jan 22, 2014 2:50am
Quoting Rap Skallion
{quote} Thank you for sharing this paper. I believe they may possibly be confirming the heavily interlocked nature of pairs with common currencies. I am tempted to build a calculator that takes the major parent currencies, measures and adds their changes, weighted with their respective sizes, and see how close it comes to the child pairs. I suspect , closer than many would believe. How's that for price predictability. I have noted how , in general, the stronger the parents push the child pair, the faster and more dependably it moves. Above a certain...
hi rap,

your taking a dive down the rabbit hole,your going to be very surprised with the goodies you will find down there.

there are certainly occasions when you will find trades which have such a high probability Its scary.

there are so many forward indications to immediate bias and market state.

im able to establish with simple calculation converged levels .these levels are very accurate and represent combined market interest converged from multiple t/f and inherent price structure.these account for many of the variables.what they don't tell me is immediate bias.i have two points of convergence,one for the buyers and one for the sellers.

so I have two price zones predicting futer price levels.what im missing is the variable of unfolding events .this is were the correlated pairs and by default cousin's
of the brothers and sisters.from this information I can more than 50 % of the time predict bias.

an example of this was yesterdays break of range on cable.which resulted in a straight 50 pip upmove.I was able to determine long bias from related gbp crosses,cable was delayed in this bias,doller related pairs were showing positive doller contrary to my bias for cable.i was able to establish the stalling point on the doller crosses and its slight weakness as doller crosses fell of there highs,my entry long on cable was clear at this point .cable quickly joined its related crosses and the exit was the lower stalling point on doller crosses.

this can be seen in the value changes on the group of pairs I study.as I mentioned in an earlier post I used to do this through excel but have found it much easer just using the values.

I would be very interested in the results you find using your maths module.

cheers
dice
  • #2,013
  • Jan 22, 2014 6:19am Jan 22, 2014 6:19am
Quoting diceman555
{quote} hi rap, your taking a dive down the rabbit hole,your going to be very surprised with the goodies you will find down there. there are certainly occasions when you will find trades which have such a high probability Its scary. there are so many forward indications to immediate bias and market state. im able to establish with simple calculation converged levels .these levels are very accurate and represent combined market interest converged from multiple t/f and inherent price structure.these account for many of the variables.what they don't...
My apologies: manually operating this "system" keeps me so busy raking in the profits I run out of energy for important things. (Shit eating grin)

A rabbit hole it is. 80% of my wins were also on the GBP crosses.

A putative calculation of the range of a child:
In quasi programming English:

FOR each parent:
IF parent is of type ++ Sum the ranges *volumes
IF parent is of type +-, or -+, take the difference of the ranges * volumes
NEXT
ADD the results and divide them by the volume of the child bar.
RETURN "child Range: " , R

I have to wonder how close and how often this algorithm would come close.

I know it is only predicting the present.
But add this simple rule: " Unless something changes, the next bar will probably be very like this one."....
  • #2,014
  • Jan 22, 2014 7:12am Jan 22, 2014 7:12am
Quoting Rap Skallion
{quote} My apologies: manually operating this "system" keeps me so busy raking in the profits I run out of energy for important things. (Shit eating grin) A rabbit hole it is. 80% of my wins were also on the GBP crosses. A putative calculation of the range of a child: In quasi programming English: FOR each parent: IF parent is of type ++ Sum the ranges *volumes IF parent is of type +-, or -+, take the difference of the ranges * volumes NEXT ADD the results and divide them by the volume of the child bar. RETURN "child Range: " , R I have to wonder...

yes unless something happens nothing will change.and if something happens what are the next 2 attraction points ?these can be calculated.and we can have a peak at that something that has not happened from money flows in or out of key currencies.quite rightly pointed out by the paper I posted were the evidence suggests certain traits can be predictive.im using the behaviour of related pairs to understand these flows and they can lead to predictive outcome in advance if only by default.

were certainly doing things differently but usin the same information
  • #2,015
  • Jan 22, 2014 9:32am Jan 22, 2014 9:32am
Quoting diceman555
{quote} yes unless something happens nothing will change.and if something happens what are the next 2 attraction points ?these can be calculated.and we can have a peak at that something that has not happened from money flows in or out of key currencies.quite rightly pointed out by the paper I posted were the evidence suggests certain traits can be predictive.im using the behaviour of related pairs to understand these flows and they can lead to predictive outcome in advance if only by default. were certainly doing things differently but usin the...

Of course -- things change when no push is on , and that is most of the time. That is where lines of expected resistance and support play a major role. If you have drawn these lines correctly, CMP will seldom cross them without being pushed... (of course the changes from asian to european to US market openings have a play. Of course, news can override all things for a while. Those items can be pushes on their own). I need to study your methods in detail more thoroughly to see if I can fit more of them in my own.

I came close to 500 pips today, and was running so fast I had to guess what my "rules" would have told me to do. A day like today brings home the necessity to have more computer assistance if you want to maximize your profits.

Parent Pair Pilot Scheme:

When no parent pushes against,and any one parent pushes with, you are in PUSH mode.
When no push is occurring (some 80% of the time) you are in Discretionary mode. ( frequently take profit at 5 pips in this mode: the maximum spread I trade).
When ANY parent pushes against (even a little bit) and you are in profit, CLOSE the trade.
When the sum of the parents pushes are against you, CLOSE the trade whether you are in profit or not. ( Don't wait for it to hit stop loss: you will only lose more money).
A news reaction is a >PUSH When , as sometimes happens, the group of bankers in N.Y. doesn't like what the Europeans are doing, a market opening can be a PUSH.

When they are all pushing together, the market can go like gangbusters( technical fiduciary term).

Now somebody ought to able to program a computer to sense these things. Basically one calculation and a set of on-off switches.

I know, I know: if not me, then who? If not Now then when? (maybe after a nice little nap).
  • #2,016
  • May 13, 2014 2:51am May 13, 2014 2:51am
Quoting diceman555
don't mention the trend word on this thread that's worth a good 100 pages of a free for all, killing field
Apologies Dice, I'm going to talk about the taboo word

Some pages ago I built random charts. They were built from the ticks return distribution of E/U to have the exact same distribution of the market. I created a random hourly chart and I added a SMA100 to it. I also added a SMA100 to the real E/U hourly. The SMA100 is supposed to represent the trend. I looked at the distribution of the 2nd difference of the SMA, that is the rate of change of the slope of the MA/trend.

The pictures show that the random trend has a slighly wider distribution of its rate of change: it wiggles more. I generated another chart to check and got twice the same result. The trend is closer to a straight line in real market data. This effect is very small but something must stabilize the trend. This is a little piece of evidence that the market can't be pure a random walk.
Attached Image (click to enlarge)
Click to Enlarge Name: trendCurvature.png Size: 9 KB
No greed. No fear. Just maths.
  • #2,017
  • May 13, 2014 3:30am May 13, 2014 3:30am
Quoting PipMeUp
{quote} Apologies Dice, I'm going to talk about the taboo word Some pages ago I built random charts. They were built from the ticks return distribution of E/U to have the exact same distribution of the market. I created a random hourly chart and I added a SMA100 to it. I also added a SMA100 to the real E/U hourly. The SMA100 is supposed to represent the trend. I looked at the distribution of the 2nd difference of the SMA, that is the rate of change of the slope of the...

That's Some amazing stuff there man, really I am impressed. Are you a mathematician or something? I hope this convinces at least some that the market is not random.
Being a trader is lonely, but being a great trader is lonelier still
  • #2,018
  • May 13, 2014 4:47am May 13, 2014 4:47am
Well this doesn't show the market isn't random. It only shows that there is at least a fraction of the move which is somewhat "known". Telling if the rest of the market movement is highly choatic yet deterministic or purely stochastic is beyond me. Anyway it wouldn't change so much for us since we know neither the equations nor the initial conditions.

http://www.physics.udel.edu/~bnikoli...stic_chaos.pdf
PS: I'm a software engineer
No greed. No fear. Just maths.
  • #2,019
  • May 13, 2014 5:34am May 13, 2014 5:34am
I think it's all how you look at it. If someone shows up at my house randomly I would be surprised because there was never any hint on before warned stipulation. But if they tell me that they are showing up at my house on friday, even if they come at midnight, no matter what time they came it wouldn't be random because I would be expecting them. And the market is the same way, we wait for the time to jump in, whenever that might be for whatever reason and more often than not you might be wrong. But if the market was random it wouldn't matter what time we jump in, it wouldn't make a difference. Probability plays a role of coarse and there are other left out stipulations. But the market does follow rules and anything that follow rules at least to me cannot be random.
Being a trader is lonely, but being a great trader is lonelier still
  • #2,020
  • May 13, 2014 5:43am May 13, 2014 5:43am
Quoting PipMeUp
Well this doesn't show the market isn't random. It only shows that there is at least a fraction of the move which is somewhat "known". Telling if the rest of the market movement is highly choatic yet deterministic or purely stochastic is beyond me. Anyway it wouldn't change so much for us since we know neither the equations nor the initial conditions. http://www.physics.udel.edu/~bnikoli...stic_chaos.pdf PS: I'm a software engineer

I have an idea that I think would be good if we can make it into a formula. I sent you a PM it's about something I have been working on but it's not quite done with a software engineer like yourself it might be able to be completed. I dont have to tell you the stakes involved if we can complete something like that. If you would be interested in hearing something that's outside the box, then reply to my PM and I will send you my ongoing research maybe you would be able to test them and come up with things that might be beyond my abilities.
Being a trader is lonely, but being a great trader is lonelier still
0 traders viewing now