EMA 25-50 Trading System (Detailed) | Page 100 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

10 Ema and 200 Ema cross system 6 replies

EMA crossover (ema 14 & 34) 4 replies

Please help code an EMA Crossover 12ema x 24 ema cross 1 reply

EMA's and MA's confiming small time correctoin ema's crossing 1 reply

EMA Close Crossing EMA High/Low Help Please! 2 replies

Attachments: EMA 25-50 Trading System (Detailed)
Exit Attachments

EMA 25-50 Trading System (Detailed)

  • #1,981
  • Jul 31, 2008 9:38am Jul 31, 2008 9:38am
Quoting Corre
I have to go out now because my parents are coming home to see me.

I'll be back this evening, I hope!

with regards to your question about whether all 4 trades will open at the same price... for that we're at the mercy of the broker and slippage. the EA will open trades 1 to 4 in sequence one after another and not concurrently. e.g. if trade 1 is opened and the price moved a pip or two at a fraction of a second before trade 2 is opened (not frequent but it happens when price surge or crash), we get a different open price. I've written EAs that opens 2 trades to manage at a time. so far, I've yet to see them having different open prices. but i can't say it won't happen and I definitely can't say for opening 4 trades. But I believe, in all likelyhood, difference if any is a matter of 1 or 2 pips.

alright. I'll have to put developments on hold at least until I get all the details about the Stop loss management from you or Toni.

It'll help if you can list out a trade example. from start till TP4 is reached. step by step as specific as possible, so I can get the logic correct.

cheers.

regards,
Zen
Programming for a better future.
  • #1,982
  • Jul 31, 2008 10:28am Jul 31, 2008 10:28am
ok. nvm I think I understand it now. should've read Toni's indicator's code a little more closely..

In fact its actually easier to code it like you described. cos that way I don't need to find a way to store the ATR value when the trade was opened.

I'll resume development again in a while. taking a short break. =)
Programming for a better future.
  • #1,983
  • Jul 31, 2008 11:49am Jul 31, 2008 11:49am
Quoting Zen_Leow
ok. nvm I think I understand it now. should've read Toni's indicator's code a little more closely..

In fact its actually easier to code it like you described. cos that way I don't need to find a way to store the ATR value when the trade was opened.

I'll resume development again in a while. taking a short break. =)

Thanks for all your hard work so far Zen, I'm really looking forward to the EA
  • #1,984
  • Jul 31, 2008 12:47pm Jul 31, 2008 12:47pm
Thanks Verto, I'm looking forward to it as well. =)
Programming for a better future.
  • #1,985
  • Jul 31, 2008 12:54pm Jul 31, 2008 12:54pm
I'll raise one more query before I sign off for the day.

I see in Toni's indicator that when a set of positions (let's say Long position) has just closed out (be it in loss or in profit), the indicator will churn out a new green arrow to Buy as long as the signal at time of close is still valid.

would you guys want that? or would you prefer alternating between buy and sell, meaning if its Buy now, the next signal to lookout for is a Sell.

the formal is good when market is trending, the latter is probably better for ranging market.

I can do it either way. your call...

alright, I'll carry on tomorrow. goodnite all!

regards,
Zen
Programming for a better future.
  • #1,986
  • Jul 31, 2008 3:09pm Jul 31, 2008 3:09pm
Quoting Zen_Leow
I'll raise one more query before I sign off for the day.

I see in Toni's indicator that when a set of positions (let's say Long position) has just closed out (be it in loss or in profit), the indicator will churn out a new green arrow to Buy as long as the signal at time of close is still valid.

would you guys want that? or would you prefer alternating between buy and sell, meaning if its Buy now, the next signal to lookout for is a Sell.

the formal is good when market is trending, the latter is probably better for ranging market.

I can do it either way. your call...

alright, I'll carry on tomorrow. goodnite all!

regards,
Zen
Hello guys, just arrived from work.

First of all, to clarify, let's call the indicator exactly that, "the indicator".
I understand why you call it mine (because you arrived to the thread recently and haven't read it much), but in it there's the work of some people, namely:
Effilang, who started the thread and inspired ronmatzkin to start coding an indicator;
ronmatzkin, who has coded all the "body" of it;
Corre who gave us some ideas for improvement;
Myself, who gave some ideas and have coded this last change.
(wouldn't want to have misunderstandments or take anyone's credit in it )

Zen:

About the changes in the ATR, I think now you understood. You don't need to store its value at the openning of a trade, just have to read its value at the moments it's needed (openning and TPs).
One thing (you might have noticed it now, reading the code): we use ATR(24).

About consecutive longs and shorts... hum... tricky question. As you say, both have their value, depending if we are trending or ranging.
Is it something hard to modify later?
Because for now I would leave it as in the indicator (possible to open consecutive longs and shorts), but would possibly try the other way later, to see the changes in performance.
  • #1,987
  • Jul 31, 2008 3:17pm Jul 31, 2008 3:17pm
One more thing,
About running the EA at a determined time of the day, I have thought of that as possible improvement to performance (have mentionned it somewhere), but as Corre said, for now keep it 24h.
The time schedule would have to be studied previously, depending on the pair in question (trading at certain hours of the day, not trading in the first hours and last hours of the trading week, etc).
  • #1,988
  • Edited 6:57pm Jul 31, 2008 4:37pm | Edited 6:57pm
about another thing, again:

are you sure it is not possible to open a trade of 0.4 at close positions by 0.1 each?

or at least to leave something configurable from outside the indicator that is a multiple of 0.4 that we can decide everytime??

for example, I'm poor and I would let it at 0.4 each with "closes" of 0.1, someone richer could set it at 4.0 or 8 or 1.2 or 1.6 or 2.4 and so on, but it has to be a multiple of 0.4

In my humble opinion with this version we can avoid the "mercy" of the broker and the slippage thing.

do you agree?

It would be great if you leave a x0.4 multiplier so we can decide if leaving it to 1, or to enlarge it.

not a micro lot trading style i suppose but this is the only way (perhaps) to avoid that crappy slippage/mercy thing. that sounded AWFUL to me.
  • #1,989
  • Edited 8:38pm Jul 31, 2008 5:23pm | Edited 8:38pm
Quoting Zen_Leow
hmm... that IS a big misunderstanding. After reading Toni's PM, I was under the impression that we only make changes to the SL levels and always in the increments of the ATR value when the trade(s) was first opened.

I'll have to rethink and work on how to get it to move the way you described. Its already night time in my timezone. looks like I won't be finishing this EA today after all.
don't worry take all the time you need, it's better if we do this thing just once even if slower than 40 times fastly

just be sure to include the last thing I said, "the lot thing" if it is possible

if it can help you I noticed that the "close lot" (the /4 lot I mean) is always 1 tenth of the multiplier.

Examples:

whole trade: 0.4
close lot: 0.1
multiplier = 1
multiplier/10 = 1/10 = 0.1 = close lot.
0.1*4tp_closures = 0.4 = whole trade

2*0.4 = 0.8 whole lot = 0.8
multiplier = 2
multiplier/10 = 2/10=0.2=close lot.
0.2*4tp_closures = 0.8 = whole trade

3*0.4 = 1.2
whole order = 1.2lots
multiplier = 3
3/10 = 0.3 = close lot
and so on

may be this can help you to set the thing properly, example: you can set a long order as

Buy x*0.4
TP=Ask Price+ATR+spread
SL=Ask Price-ATR-spread

TP1 reached
close position for 1/10*x lots
set SL to entry (Ask price+spread) for the last 3 lots
and set new TP as the new ATR

that would be perfect in my opinion, it can avoid the whole mercy/slippage problem and can leave the choice of the size of the orders to everyone of us freely.

Can't think to a better way to close the trades

it is very easy and cool and probably easy to code too.

leave eveyone free to set their multiplier, then:

entry order = Multiplier * 0.4
exit quarter order = Multiplier * 0.1

how cool is this?
  • #1,990
  • Jul 31, 2008 6:12pm Jul 31, 2008 6:12pm
Quoting tonirai
Hello guys, just arrived from work.

First of all, to clarify, let's call the indicator exactly that, "the indicator".
I understand why you call it mine (because you arrived to the thread recently and haven't read it much), but in it there's the work of some people, namely:
Effilang, who started the thread and inspired ronmatzkin to start coding an indicator;
ronmatzkin, who has coded all the "body" of it;
Corre who gave us some ideas for improvement;
Myself, who gave some ideas and have coded this last change.
(wouldn't want to have misunderstandments or take anyone's credit in it )
by the way, thank you for these words. very kind from you
  • #1,991
  • Edited 10:07pm Jul 31, 2008 9:50pm | Edited 10:07pm
Quoting tonirai
One more thing,
About running the EA at a determined time of the day, I have thought of that as possible improvement to performance (have mentionned it somewhere), but as Corre said, for now keep it 24h.
The time schedule would have to be studied previously, depending on the pair in question (trading at certain hours of the day, not trading in the first hours and last hours of the trading week, etc).
not to worry, I code time management in all my EAs. you will get to choose a start time and number of hours to trade. and if you set number of hours to 24, it will disregard start time and run for as long as the EA is operational. I think this is best for everyone.

regards,
Zen

***EDIT***
come to think of it... I can allow an Input option that sets whether or not to use consecutive signals or alternating signals... ok I'll do that. just thinking out loud now. =)
Programming for a better future.
  • #1,992
  • Jul 31, 2008 9:54pm Jul 31, 2008 9:54pm
Quoting tonirai
About consecutive longs and shorts... hum... tricky question. As you say, both have their value, depending if we are trending or ranging.
Is it something hard to modify later?
Because for now I would leave it as in the indicator (possible to open consecutive longs and shorts), but would possibly try the other way later, to see the changes in performance.
fyi, I've coded both methods in. default is set to allow consecutive signals. changing to alternating signals is only a matter of uncommenting a few lines of codes and recompile the EA.

regards,
Zen
Programming for a better future.
  • #1,993
  • Jul 31, 2008 9:57pm Jul 31, 2008 9:57pm
Quoting Corre
about another thing, again:

are you sure it is not possible to open a trade of 0.4 at close positions by 0.1 each?

or at least to leave something configurable from outside the indicator that is a multiple of 0.4 that we can decide everytime??

for example, I'm poor and I would let it at 0.4 each with "closes" of 0.1, someone richer could set it at 4.0 or 8 or 1.2 or 1.6 or 2.4 and so on, but it has to be a multiple of 0.4

In my humble opinion with this version we can avoid the "mercy" of the broker and the slippage thing.

do you agree?

It would be great if you leave a x0.4 multiplier so we can decide if leaving it to 1, or to enlarge it.

not a micro lot trading style i suppose but this is the only way (perhaps) to avoid that crappy slippage/mercy thing. that sounded AWFUL to me.
Corre, I didn't say it can't be done, I'm just saying its better to use the 4 positions method. I think its best I explain what I meant by opening 4 positions. Should've done so earlier but I thought I was going to finish the EA sooner and the Input's description would explain it all...

I incorporate MoneyManagement in all my EAs. you'll get to set whether you want MoneyManagement on or off, if off, use fixed Lots (configurable). if on, you can set a certain RiskPercent(also configurable) of Account Balance for each trade which will calculate the optimum lot size using the system derived stop loss pips. Then there's a SplitVolume option to determine if you'd want to split the derived Lot size among the 4 positions, or have them all use that derived lot size.

but note that all positions' lot size is also governed by minimum and maximum lots allowed by the broker. This is vital as even IF we were to go the route of partial closing of positions (which EA is not), those partial closes must also be within the limits of the broker's minimum and maximum lot size.
So all things considered, I think opening 4 positions and setting their relative TPs has somewhat an advantage over opening 1 position and do partial closing.

programatically, I feel its actually harder to code the handling of 1 position and do partial closing. EAs in general are stateless programs... when the EA is reinitialised, it has no idea how a previously opened trade has developed. and I think the EA must somehow be able to know at which stage the trade is in after the event of a PC reset or deinitialise, when the EA is forced to reinitialise. I'm trying to allow as much as possible... for the EA to be unaffected by such scenarios.

and also if your EA cannot be used (perhaps PC down, power failure, etc) while the trade has just opened. price rose to TP3 level, then fell all the way down... with 1 position and no EA to handle stoploss management, you will suffer a total loss. with 4 positions, at least 3 will have reached their initial TPs and closed off leaving the last one with a loss.

Handling 4 positions, is not without its share of difficulties, but that's something I believe I can resolve. Both methods have their pros and cons and it just seems to me that using "4 positions" outweighs "1 position" by quite a bit. So that's the path I'm going with the EA. and rest assured the 4 positions' lot sizes are done just as how it would've been if I were to code "1 position and partial closing" method.

the "mercy of the broker and the slippage thing" only refers to the opening price which (I hope) in all likelihood would be a matter of 1 or 2 pips difference. if I were to weigh this against disaster recovery as I mentioned above... I choose disaster recovery.

I hope this clears your doubts.

Having said all these, I must say I'm very excited at how it'll all turn out. =)

regards,
Zen
Programming for a better future.
  • #1,994
  • Jul 31, 2008 10:12pm Jul 31, 2008 10:12pm
Quoting Corre
don't worry take all the time you need, it's better if we do this thing just once even if slower than 40 times fastly

just be sure to include the last thing I said, "the lot thing" if it is possible

if it can help you I noticed that the "close lot" (the /4 lot I mean) is always 1 tenth of the multiplier.

Examples:

whole trade: 0.4
close lot: 0.1
multiplier = 1
multiplier/10 = 1/10 = 0.1 = close lot.
0.1*4tp_closures = 0.4 = whole trade

2*0.4 = 0.8 whole lot = 0.8
multiplier = 2
multiplier/10 = 2/10=0.2=close lot.
0.2*4tp_closures = 0.8 = whole trade

3*0.4 = 1.2
whole order = 1.2lots
multiplier = 3
3/10 = 0.3 = close lot
and so on

may be this can help you to set the thing properly, example: you can set a long order as

Buy x*0.4
TP=Ask Price+ATR+spread
SL=Ask Price-ATR-spread

TP1 reached
close position for 1/10*x lots
set SL to entry (Ask price+spread) for the last 3 lots
and set new TP as the new ATR

that would be perfect in my opinion, it can avoid the whole mercy/slippage problem and can leave the choice of the size of the orders to everyone of us freely.

Can't think to a better way to close the trades

it is very easy and cool and probably easy to code too.

leave eveyone free to set their multiplier, then:

entry order = Multiplier * 0.4
exit quarter order = Multiplier * 0.1

how cool is this?
yes its cool
BUT... I kinda think its too much math for something simple.
so... opening 4 trades and closing them off 1 at a time is easier.
Programming for a better future.
  • #1,995
  • Jul 31, 2008 11:51pm Jul 31, 2008 11:51pm
yes but the slippage thing is important... I wouldn't understimate it, isn't that system easy to add? It's just a matter of some moltiplications, isn't it?
  • #1,996
  • Aug 1, 2008 2:11am Aug 1, 2008 2:11am
Quoting Corre
yes but the slippage thing is important... I wouldn't understimate it, isn't that system easy to add? It's just a matter of some moltiplications, isn't it?
I'm afraid its not that simple. and I've explained the reason in my long post above. I'm just trying to play safe here... just think about disaster recovery and you'll get what i mean.

tell you what... I'll finish up my 4 position version. and IF I have time, I'll code a separate version that only works on 1 position. how about that?
Programming for a better future.
  • #1,997
  • Aug 1, 2008 2:56am Aug 1, 2008 2:56am
anything would be cool, sure.

if you find the time to work on the single order one, please consider this type of trading:

(multiplier =z)
long order:
buy z*0.4
Set SL = Ask Price - ATR + spread (pay attention, it's + spread, not -)
----> with no TP <----

AT THE SAME TIME:

place a limit order that triggers at
Ask Price the (1 of before) + ATR + spread
and order it to sell 0.1*z

then once the limit order triggers we have to move the SL of the main trade at BE and place another limit order at
TP1 + ATR

and you already know all the rest

Imho this way is pretty easy to manage anything.
  • #1,998
  • Aug 1, 2008 4:07am Aug 1, 2008 4:07am
Quoting Corre
anything would be cool, sure.

if you find the time to work on the single order one, please consider this type of trading:

(multiplier =z)
long order:
buy z*0.4
Set SL = Ask Price - ATR + spread (pay attention, it's + spread, not -)
----> with no TP <----
this is the first time I've heard of this kinda stop loss. what's your reason for using + spread for a long order?? as I look at it, you actually want a smaller stoploss than what ATR has given.

standard practise is
SL = Ask Price - ATR - (Spread * Point)
Programming for a better future.
  • #1,999
  • Aug 1, 2008 5:17am Aug 1, 2008 5:17am
anyway if we're using the ask price, that is the price in which we enter, we don't need to take care of any spread.

the correct SL is just Ask price - ATR(24). isi
  • #2,000
  • Aug 1, 2008 6:53am Aug 1, 2008 6:53am
no... i think u confused SL w TP calculation. u have to include spread in the calculation of stoploss cos the stoploss will activate once the BID reach the stoploss level u set. remember the bid price is lower than ask, that means for a long order, u're off with a lower starting point, u have to factor this difference in.

we don't need to include spread into TP calculation because like i said, u're already off with a lower start, so when TP is concerned spread is already factored in once the position opens.
Programming for a better future.
9 traders viewing now