Forex for Dummies

[Cartman]

i know you did. hence my comment on having all my money on black

my stop loss is just below 71. had a lucky life on that bounce. be interesting to see where it ends up.

i knew u knew -----

ps no need to tell me that u knew that i knew that u knew ----

cause i already know that!

hope 71 holds for ya -- looked like a fair bit of buying came in down there so the next low cycle will prob tell the story

 

[lesm]

For those of you who say news items spike the markets:
From Money Morning:
In their recent paper titled Stock price jumps: news and volume play a minor role, physicists Armand Joulin, Augustin Lefevre, Daniel Grunberg and Jean-Philippe Bouchaud found no correlation between news items and the responses of stocks. They analysed the news feeds from both Dow Jones and Reuters (the major sources of information for financial analysts and journalists) and examined the correlation between hundreds of instruments and some 90,000 news items over a two-year period.

Their conclusion was there was no link between jumps in instrument pricing and news items. Most changes were not directly attributable to any news item at all and the majority of news items caused no change in instrument pricing at all.

comments please?
I thought Gold would go up not down after the Gaza fighting.

It's all a figment of the imagination

They were not watching the currency markets when the Lehman's news came out on Reuters or the subsequent updates, but then they are talking about stocks.

Sounds like we can ignore the news Bentrod, NFP, PPI, rate changes, etc., will have no affect.

 

[lesm]

Norman....good to see that you have been doing well.

Thought I would throw up some testing stats as to the potential of the basic method that Norman is using, with some refinements.

The fundamental basis is the same, but:

1. The stoch settings are different
2. Scaling in and out is being applied
3. Additional supporting pattern analysis is being used
4. Daily pivots are being used, as reference points.

This was a refinement to an earlier system that I was having some consistency issues with. Took a few days to adapt to it, but it clicked into gear. The tests were conducted during August and September last year and the system is now being used live.

All individual trades are 1 std contract.

The main sessions being traded are the London and London / New York overlap.

Trading is restricted to four pairs: aud/usd, eur/gbp, eur/usd and gbp/usd.

Cheers.

 

[Stormin_Norman]

what's the conclusion?

 

[CanOz]

That looks like quite a high win rate. Do you have a cumulative equity curve as well, not just the daily trades?

Cheers,


CanOz

 

[hotbmw]

so guys does that mean for a mere $1000 micro account i can trade 50 times that = $50,000.

if i got it wrong and the trade went the other way by say 3% before closing out i owe $1500 even though i started with $1000 only? is that right?

or if i made a good buy and i closed out with a 5% gain, i would make $2500 with only an initial outlay of $1000.

is that how it works?

is there any interest payments or monthly fees of any sort if i hold a trade for months without closing out? any fees of any osrt?

 

[MS+Tradesim]

Hi Les,

Those stats are interesting. I've been testing Norman's idea too but I get nothing like your results. I must be misunderstanding something very simple.

 

[Stormin_Norman]

so guys does that mean for a mere $1000 micro account i can trade 50 times that = $50,000.

if i got it wrong and the trade went the other way by say 3% before closing out i owe $1500 even though i started with $1000 only? is that right?

or if i made a good buy and i closed out with a 5% gain, i would make $2500 with only an initial outlay of $1000.

is that how it works?

is there any interest payments or monthly fees of any sort if i hold a trade for months without closing out? any fees of any osrt?

you can only lose what u have deposited. youll get a margin call and have your positions closed before then.

you sound like youre looking at overleveraging. only lose 5% max per trade.

u do get swap rates for long term trades. which can also 'cost' you if youre on the lower interest rate side of the trade.

 

[Stormin_Norman]

Hi Les,

Those stats are interesting. I've been testing Norman's idea too but I get nothing like your results. I must be misunderstanding something very simple.

we're finding difficulties with the automation too; not having human intuition as part of the computer trading is difficult.

we're looking at the slope of the MA line atm to see if that helps some.

 

[hotbmw]

you can only lose what u have deposited. youll get a margin call and have your positions closed before then.

you sound like youre looking at overleveraging. only lose 5% max per trade.

u do get swap rates for long term trades. which can also 'cost' you if youre on the lower interest rate side of the trade.

hey stormin. thanks for your reply mate.

so say i open an account for $5000. instead of investing $250,000 i invest $50,000.
not that i would do this but does that mean my trade would need to go bad by the tune of 10% before im closed out?

where can i find info on swap rates for IB? is it expensive?

 

[MS+Tradesim]

we're finding difficulties with the automation too; not having human intuition as part of the computer trading is difficult.

we're looking at the slope of the MA line atm to see if that helps some.

Be very careful with using slope/degrees. It's arbitrary based on the time/price scale. eg. It might "look" like a 20 degree slope but zoom in or out on the chart and the angle will change. You may already know that and have a constant to base it against, but just thought I'd point it out in case you didn't already know.

One way around that problem might be to use a rate of change of an MA or rate of change of a linear regression slope.

 

[Stormin_Norman]

Be very careful with using slope/degrees. It's arbitrary based on the time/price scale. eg. It might "look" like a 20 degree slope but zoom in or out on the chart and the angle will change. You may already know that and have a constant to base it against, but just thought I'd point it out in case you didn't already know.

One way around that problem might be to use a rate of change of an MA or rate of change of a linear regression slope.

that's what i meant. the 5 period slope of the MA line.

 

[lesm]

That looks like quite a high win rate. Do you have a cumulative equity curve as well, not just the daily trades?

Cheers,


CanOz

Agreed, Keep in mind it's only over 375 trades (with a win rate of 73%) needs to be over at least a 1000 or more to see if it holds up through time.

We (myself and two other colleagues) were testing a number of approaches at the time and this was one of the more positive ones.

Haven't calculated the curve, but happy to send you a copy of the raw data if you want to analyse it. I use Oanda for testing purposes only, so you will be familiar with the format.

Using a combination of price/patterns/stochs appears to strengthen it, rather than the just the stochs in isolation. The scaling in and out with additional contracts, when it is warranted, does have an accelerating and positive effect.

I have seen people try to use stochs or other indicators in isolation and a lot of them fail long term. That doesn't mean particular individuals cannot do it. Some people just seem to be able to interpret and apply particular indicators in a way that works for them. There may be some subconcious or intuitive factors coming into play.

 

[Stormin_Norman]

do u use support/resistance?

how do u look to program that?

 

[lesm]

tech/a,

You may be interested in taking out a free subscription to Currency Trader magazine:

http://www.currencytradermag.com/subscribe/index.php

It provides a range of information on currencies, as well as statistics that you may find of interest.

Can drop 18 months worth of back issues on a CD and mail them to you if you like. Too big to email.

Cheers.

 

[Stormin_Norman]

that mag would be good if it wasnt 2/3rds advertisements.

 

[BentRod]

Lesm,
Are those stats from Oanda trades?

How did you get them into excel?

cheers

 

[Cartman]

Using a combination of price/patterns/stochs appears to strengthen it, rather than the just the stochs in isolation. The scaling in and out with additional contracts, when it is warranted, does have an accelerating and positive effect.

I have seen people try to use stochs or other indicators in isolation and a lot of them fail long term. That doesn't mean particular individuals cannot do it. Some people just seem to be able to interpret and apply particular indicators in a way that works for them. There may be some subconcious or intuitive factors coming into play.

hey Les (and Norm) as ive said b4 i know nothing about coding etc etc ---- but re the entries via Stochastics which eventually fail scenario ---

there must ?? be a way to filter that problem out by ---- for eg ---- the entry on the low Stoch position is only taken when on a higher time frame the short MA is above the longer MA etc etc ---- if u get my drift -- i drift a lot lol ----

the higher time frame may then be filtered by yet a higher time frame (if possible when coding ??) --- to eliminate false entries ---

kinda all depends on the time frame each person is trading in as well ---

that is why i forward test my ideas --- correct coding for back testing must be a nightmare

 

[Stormin_Norman]

we look at a higher timeframe for exits. could be an interesting thought for entries too. i will stew on it more.

has anyone successfully coded resistance/support?

 

[Cartman]

we look at a higher timeframe for exits. could be an interesting thought for entries too. i will stew on it more.

has anyone successfully coded resistance/support?

yeah i think entries prob deserve a lot more attention than many give creedence to ---- after all -- the entry is the most vulnerable part of the trade ---- thats why i like to scale in -- which i know u do as well Norm --- but how the heck u guys code this stuff is beyond me ---- good luck with it