Trading The "Normal State" | Page 5 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Broker Challenge "NDD/STP" vs. True "ECN" and "Mkt Mkr" broker talk 46 replies

Dealing with "Invalid Price", "Requote", "Server Busy" in MT4 3 replies

What do you consider a "normal" ping on MT4? 1 reply

Forex spread betting versus "normal" trading. 1 reply

It’s our job to trade "Futures" not "Histories", by Joe Ross 7 replies

Attachments: Trading The "Normal State"
Exit Attachments

Trading The "Normal State"

  • #81
  • Oct 11, 2007 12:40am Oct 11, 2007 12:40am
Only sort of....The relationships hold if everything is static. The fun is when something unusual happens. Suppose on a nice quite Sunday night, somebody decides he just has to have to buy a few billion Euros using Yen.

This may go back to what Vikrant said-banks want dollars because it's the world's reserve currency so in this case the Yen are converted into Dollars first and those Dollars are converted to Euros.

That seems to be what happens when carry trades unwind with equities. High yielders are converted to Dollars and Dollars are converted into Yen.
  • #82
  • Oct 11, 2007 3:45am Oct 11, 2007 3:45am
I have read this interesting thread and as I am a carry trader (I trade it of course on a real account) I wanted to react to some figures I saw like 125% and 80% of return.
I will not describe here how I trade but it is not only buy and hold, it is a bit more complicated and more effective (at least I think so), but I would like to share my experience to give an idea with more reasonable numbers.

The numbers like 125% or 80% are calculated as some of you noticed it with far too much leverage (specially considering GBPJPY volatility) and your account would have been wipped out either in February/March or in August/September.

For me a good return for the carry trading strategy is around 50% pa and from this 50% clearly less than half comes purely from the swap and the rest comes from the pip profits because of the trend of the pairs.
If you would just buy and hold, you would probably not get more than 25% pa in order to have reasonable DD (I imagine 15% from swap and 10% from pip profits).
On good months (like now just after the drop) you can easily make 10 - 15% a month but it is not the average return.
  • #83
  • Oct 11, 2007 6:51am Oct 11, 2007 6:51am
Quoting NewstraderFX
Only sort of....The relationships hold if everything is static. The fun is when something unusual happens. Suppose on a nice quite Sunday night, somebody decides he just has to have to buy a few billion Euros using Yen.

This may go back to what Vikrant said-banks want dollars because it's the world's reserve currency so in this case the Yen are converted into Dollars first and those Dollars are converted to Euros.

That seems to be what happens when carry trades unwind with equities. High yielders are converted to Dollars and Dollars are converted into Yen.
Pretty much, except that in that role the USD is acting as a vehicle currency - there's a number of good papers available from the FRB on such, this one (http://129.3.20.41/eps/if/papers/0111/0111001.pdf ) goes into a bit of background, though it dates from the 90's.

OTOH when it is the ccy that is being held widely as a store of wealth, it's a reserve ccy. (good article at http://en.wikipedia.org/wiki/Reserve_currency )

In the last article it's interesting to note the rise of the euro as a reserve ccy since 2000.
  • #84
  • Oct 11, 2007 8:23am Oct 11, 2007 8:23am
Quoting jlpi
The numbers like 125% or 80% are calculated as some of you noticed it with far too much leverage (specially considering GBPJPY volatility) and your account would have been wipped out either in February/March or in August/September.
I pointed that out back on page 1 in the third post of this thread. But, there's no fun in injecting real-life scenarios into our time here at fantasy camp.
  • #85
  • Oct 11, 2007 8:37am Oct 11, 2007 8:37am
Quoting HalifaxCB
The only problem is is that the robots can only guess at what the demand for the other legs might be; from what I understand it's based on things like kalman filters & bayesian analysis & correlations, tobots in the long run aren't real good at open questions.
hmm, i dont think its that complicated. all a broker needs is a few equations to tell where the market should be for each pair. for instance, lets say the broker has the following equations displayed in realtime (with real prices)...

GBP/JPY = GBP/USD x USD/JPY
EUR/CHF = EUR/USD x USD/CHF
EUR/GBP = EUR/USD divided by GBP/USD
GBP/CAD = GBP/USD x USD/CAD
AUD/NZD = AUD/USD divided by NZD/USD

we know that all of these equations should hold TRUE (otherwise there is arb opportunity). so all you would need is some software that alerts when an equation is FALSE. and when a false signal is given, the software adjusts the price of each pair until they all hold true again.

now, if you an arb trader (instead of broker), the software would alert of a FALSE and then make the appropriate trade...in other words, robot arbitrage. it can all be done by analyzing a handful of equations like the ones above. arb traders and brokers use the same exact equations to find out where the market should be!
Relax and be happy.
  • #86
  • Oct 11, 2007 8:37am Oct 11, 2007 8:37am
Thanks merlin for the reply, very interesting these bots. So that's why there are so many automated "systems" in the market - I always thought there could not have been so many winning mechanical strategies, but this makes sense to use a program for.

I guess it comes down to who gets the quotes first, and people would probably want an ECN for this sort of data.

About the actual idea the poor guy was trying to put forward, I suggest we wait 6 months and come back with the price of GBPJPY then. And if you can just survive with leverage for that long, newstrader, let alone make 80% of an amount invested in buying GBP now, then you da man.
Virtue finds and chooses the mean. Aristotle, Ethica Nichomachea
  • #87
  • Oct 11, 2007 8:42am Oct 11, 2007 8:42am
Quoting merlin
now, if you an arb trader (instead of broker), the software would alert of a FALSE and then make the appropriate trade...in other words, robot arbitrage. it can all be done by analyzing a handful of equations like the ones above. arb traders and brokers use the same exact equations to find out where the market should be!
I would think that the spreads and the lack of sophistication available to a retail trader would make this nearly impossible to trade. As the market becomes more and more efficient these opportunities begin to evaporate.
  • #88
  • Oct 11, 2007 8:43am Oct 11, 2007 8:43am
Quoting NewstraderFX
Every currency equation there is can be re-written in terms of the dollar
every currency equation there is can be re-written in terms of ANY currency. USD, NZD, EUR, it doesnt matter which it is, the equations work just the same.
Relax and be happy.
  • #89
  • Oct 11, 2007 8:44am Oct 11, 2007 8:44am
Quoting philmcgrew
I would think that the spreads and the lack of sophistication available to a retail trader would make this nearly impossible to trade. As the market becomes more and more efficient these opportunities begin to evaporate.
oh yeah, those opportunities were gone like 100 years ago LOL

but they still appear during news releases. thats why i tried to capitalize on those times. and like i said, i had some success but im not trading it anymore if that tells ya anything.
Relax and be happy.
  • #90
  • Oct 11, 2007 8:58am Oct 11, 2007 8:58am
Quoting merlin
every currency equation there is can be re-written in terms of ANY currency. USD, NZD, EUR, it doesnt matter which it is, the equations work just the same.
How-I can't figure out how to do this
  • #91
  • Oct 11, 2007 9:13am Oct 11, 2007 9:13am
Quoting NewstraderFX
How-I can't figure out how to do this
Yes NFX...what Merlin has written is correct. However if one is transacting on the ground level, then one cannot escape the USD leg.

Suppose I want to change my currency from INR to EUR, then my moneychanger converts INR first into USD, and then he converts USD into EUR. That's the way the world is structured, as the USD is the only foreign accepted currency in some countries.

This relationship would only change if USD is no longer the Reserve currency of the world, like it happened to the pound 30-40 years ago, when USD became the Reserve currency.

USD is the Reserve currency, and that is why all the Reserves of all the central banks (Asian and European) are declared in terms of USD and Local currency.

Trade Wise, Trade Well!
  • #92
  • Oct 11, 2007 9:15am Oct 11, 2007 9:15am
Quoting NewstraderFX
How-I can't figure out how to do this
give me an example of one you want to do and i will bring you through it.

Quoting Vikrant Goyal
That's the way the world is structured, as the USD is the only foreign accepted currency in some countries.
aha yes, thats the answer. i was saying "thats just the way it is", but your right, this is the specific reason.

i love the TR
Relax and be happy.
  • #93
  • Oct 11, 2007 9:24am Oct 11, 2007 9:24am
Quoting Plutonite
I always thought there could not have been so many winning mechanical strategies, but this makes sense to use a program for.
these are the type of mechanical systems that actually work. when you hear about guys making big money with mecahnical systems, this is the sort of stuff they are doing. mostly it is "risk-arb", which is of course arbitrage with a little bit of risk (beyond normal execution risk). the reason most have resorted to risk-arb is because all the pure arb opportunities are long gone with the advent of the computer.

the mechanical systems that buy/sell based on indicators and all that, the kind we see all around the forums...well lets just say im a skeptic i dont believe in any strategy that can be definitively "backtested" using a program like MT or TS.
Relax and be happy.
  • #94
  • Oct 11, 2007 9:37am Oct 11, 2007 9:37am
Quoting merlin
hmm, i dont think its that complicated. all a broker needs is a few equations to tell where the market should be for each pair. for instance, lets say the broker has the following equations displayed in realtime (with real prices)...

GBP/JPY = GBP/USD x USD/JPY
EUR/CHF = EUR/USD x USD/CHF
EUR/GBP = EUR/USD divided by GBP/USD
GBP/CAD = GBP/USD x USD/CAD
AUD/NZD = AUD/USD divided by NZD/USD
For static (or instantaneous) prices, this is true. But prices aren't static, and trades take a finite time to execute. For me, it's not a significant deal, because my acct is too small to take advantage of it. For a bank or a big trader (particularly algorithmic traders), it is significant, and you have to look at it in terms of the actual uncertainty (i.e. spread) of the related cross trades.
  • #95
  • Oct 11, 2007 10:03am Oct 11, 2007 10:03am
Quoting HalifaxCB
For static (or instantaneous) prices, this is true. But prices aren't static, and trades take a finite time to execute.
yeah well that "lag" is why arb guys are in business. and they do a very good job of keeping prices in line. punch in some prices at any given time, the equations will be true 99.99% of the time.
Relax and be happy.
  • #96
  • Oct 11, 2007 4:08pm Oct 11, 2007 4:08pm
Quoting HalifaxCB
Lol, I think we are working along the same lines here..The only problem is is that the robots can only guess at what the demand for the other legs might be; from what I understand it's based on things like kalman filters & bayesian analysis & correlations, tobots in the long run aren't real good at open questions. One of these days we might get lucky and get the equivalent of the portfolio insurance breakdown that got the stock markets in the late 80's. Keep liquid
Now the Kalman filter should provide a good basis for market scalping. The filter is intended to take unsynchronised, error ridden or at least occasionally innaccurate, data and figure out what the offsets, time delays and current state of anything might be. It should also be able to suss the trading lag and determine a few useful things - such as where your rates are likely to be within 5 or 10 minutes, and more particularly where your rates will be if you buy 10-billion euros using palm-oil.

Unfortunately I don't think any of us has access to the data in order to start working on the filter; and if I had 10-billion euros worth of palm-oil I don't think I would bother.
Minesweeper target: 284
  • #97
  • Oct 11, 2007 4:40pm Oct 11, 2007 4:40pm
Quoting Primate
Now the Kalman filter should provide a good basis for market scalping. The filter is intended to take unsynchronised, error ridden or at least occasionally innaccurate, data and figure out what the offsets, time delays and current state of anything might be. It should also be able to suss the trading lag and determine a few useful things - such as where your rates are likely to be within 5 or 10 minutes, and more particularly where your rates will be if you buy 10-billion euros using palm-oil.
As long as everything stays "normal", . It's funny, I play with them a lot, but I've never tried it on something explicitly "hard", like a Cauchy distribution. Maybe I should.

Quote
Unfortunately I don't think any of us has access to the data in order to start working on the filter; and if I had 10-billion euros worth of palm-oil I don't think I would bother.

The basic filter is pretty easy, for most sane languages, and doable in metatrader , but not something I advise (I am happy with it re. just price data, though analyzing things like tick rates could be interesting...)...Generally I find it a better part of a Bayesian classification scheme than (say) moving averages, or LSQ fitting....(am I showing my biases here?)
  • #98
  • Oct 11, 2007 5:47pm Oct 11, 2007 5:47pm
The "Fast Arb" thingy is something I tried a while back.

http://www.forexfactory.com/showpost...40&postcount=4

It worked occasionally but the Spread on the Crosses reduced the number of Real opportunities.

btw. the ScrSht is a chopped XL File which was rather large and had a tonne more info to spew and decipher.
TR
  • #99
  • Oct 31, 2007 9:37am Oct 31, 2007 9:37am
I thought I would update this.

As you can see, the nornal state of the economy is expansion. I've been riding the up and downs in GBP/JPY, collecting my interest everyday. It does get scary at times but those times also represent buying oppurtunities.
  • #100
  • Oct 31, 2007 9:43am Oct 31, 2007 9:43am
Quoting NewstraderFX
I thought I would update this.

As you can see, the nornal state of the economy is expansion. I've been riding the up and downs in GBP/JPY, collecting my interest everyday. It does get scary at times but those times also represent buying oppurtunities.
The GBPJPY is actually about the same as it was back on October 7th when you made the first post in this thread. The SPX is actually lower. How that supports your lame-ass expansion theory is beyond me.

If your argument is that the economy has expanded since the great depression then a bi-montly update is just noise.
2 traders viewing now