scooby-doo BOJ JPY pairs trading robot | Page 295 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Trading correlation pairs by using the other pairs 843 replies

Export All pairs or selected pairs to .csv with script 3 replies

Pairs of Currency Pairs 4 replies

Trending Pairs / Ranging Pairs 0 replies

Robot trading 120 replies

Attachments: scooby-doo BOJ JPY pairs trading robot
Exit Attachments

scooby-doo BOJ JPY pairs trading robot

  • #5,881
  • Feb 13, 2010 4:21am Feb 13, 2010 4:21am
Quoting watube
Steve, Scoobs,
This is just info. I was just now looking at the 2 digit jpy pair trades to see why they are not getting closed. I have seen what the problem is. It's the way the bot calculates TP and SL for the 2 digit pairs. Eg 20 pip TP is placed on these pairs by the bot as 200 pips and 80 pip SL as 800 pip SL.
The TP and SL is placed correctly by the bot on the other (not 2 digit) pairs. I will just exclude these 2 digit jpy pairs for now.
I have not had this problem and cannot see how it arose. I have the bot on an Alpari demo (5 digits) and live on eglobal (4 digits). All tp's were correctly placed. It does not matter how hard I glare at the code, I cannot see anything wrong.

You did have the robot on a GJ chart didn't you?

  • #5,882
  • Feb 13, 2010 4:30am Feb 13, 2010 4:30am
Quoting SteveHopwood
I have not had this problem and cannot see how it arose. I have the bot on an Alpari demo (5 digits) and live on eglobal (4 digits). All tp's were correctly placed. It does not matter how hard I glare at the code, I cannot see anything wrong.

You did have the robot on a GJ chart didn't you?

I dont understand this either but it looks like some brokers have 5 digits for some JPY pairs and 4 digits for other JPY pairs on the same account.

I would seriously question why they are doing this as it can only be to catch EA's out etc.

Scoobs.
  • #5,883
  • Feb 13, 2010 4:35am Feb 13, 2010 4:35am
Quoting scooby-doo
I dont understand this either but it looks like some brokers have 5 digits for some JPY pairs and 4 digits for other JPY pairs on the same account.

I would seriously question why they are doing this as it can only be to catch EA's out etc.

Scoobs.
The fix is easy. Do you want me to do it?

  • #5,884
  • Feb 13, 2010 4:37am Feb 13, 2010 4:37am
Quoting SteveHopwood
The fix is easy. Do you want me to do it?

Yes,

I already added it to the Yen Trader EA but not my others.
  • #5,885
  • Feb 13, 2010 4:55am Feb 13, 2010 4:55am
Attached.

Attached File(s)
File Type: mq4 Scooby 20EMA-200EMA-RSI3.mq4   8 KB | 286 downloads
  • #5,886
  • Feb 13, 2010 5:08am Feb 13, 2010 5:08am
Quoting SteveHopwood
Attached.

There is a very simple change to cut out bad trades and therefore DD.

At the moment we have....

double EMA_020 = iMA(Pairs[b],H04,020,0,MODE_EMA,PRICE_CLOSE,0);
double EMA_200 = iMA(Pairs[b],H04,200,0,MODE_EMA,PRICE_CLOSE,0);

And if you change it too....

double EMA_020 = iMA(Pairs[b],H01,020,0,MODE_EMA,PRICE_CLOSE,0);
double EMA_200 = iMA(Pairs[b],H04,200,0,MODE_EMA,PRICE_CLOSE,0);

Scoobs
  • #5,887
  • Feb 13, 2010 5:13am Feb 13, 2010 5:13am
Quoting scooby-doo
There is a very simple change to cut out bad trades and therefore DD.

At the moment we have....

double EMA_020 = iMA(Pairs[b],H04,020,0,MODE_EMA,PRICE_CLOSE,0);
double EMA_200 = iMA(Pairs[b],H04,200,0,MODE_EMA,PRICE_CLOSE,0);

And if you change it too....

double EMA_020 = iMA(Pairs[b],H01,020,0,MODE_EMA,PRICE_CLOSE,0);
double EMA_200 = iMA(Pairs[b],H04,200,0,MODE_EMA,PRICE_CLOSE,0);

Scoobs
Done.

Attached File(s)
File Type: mq4 Scooby 20EMA-200EMA-RSI3.mq4   8 KB | 823 downloads
  • #5,888
  • Feb 13, 2010 5:16am Feb 13, 2010 5:16am
Quoting SteveHopwood
Ehup guys. The end of another trading week, and a good one. Here is where I am, in order of current favourites:

 

  1. .................. Cannot wait for next week.

Have a great weekend everyone.

Hi, Like you, I cannot wait the next week to see the bots working.
But this week-end will be nice to add some features to my bot.
Here is the equity / balance graph for friday.
Approx. 500 trades with the 5 modes, nice for further analysis.
Attached Image
  • #5,889
  • Feb 13, 2010 5:22am Feb 13, 2010 5:22am
Quoting scooby-doo
There is a very simple change to cut out bad trades and therefore DD.

At the moment we have....

double EMA_020 = iMA(Pairs,H04,020,0,MODE_EMA,PRICE_CLOSE,0);
[b]double EMA_200 = iMA(Pairs,H04,200,0,MODE_EMA,PRICE_CLOSE,0);

And if you change it too....

[b]double EMA_020 = iMA(Pairs[b],H01,020,0,MODE_EMA,PRICE_CLOSE,0);
double EMA_200 = iMA(Pairs[b],H04,200,0,MODE_EMA,PRICE_CLOSE,0);

Scoobs
Scoob, what do think about someting like :
if EMA20[H1]>EMA200[H1] && EMA20[H4]>EMA200[H4]
personnaly in my bot I have used :
if EMA20[M15]>EMA200[M15] && EMA20[H1]>EMA200[H1]
  • #5,890
  • Feb 13, 2010 5:25am Feb 13, 2010 5:25am
Quoting maxou888
Scoob, what do think about someting like :
if EMA20[H1]>EMA200[H1] && EMA20[H4]>EMA200[H4]
personnaly in my bot I have used :
if EMA20[M15]>EMA200[M15] && EMA20[H1]>EMA200[H1]
if EMA20[H1]>EMA200[H1] && EMA20[H4]>EMA200[H4]

Yes, this will give you added protection.
  • #5,891
  • Feb 13, 2010 5:33am Feb 13, 2010 5:33am
this is your last suggestion
EMA20[H1] > EMA200[H4] (I put EMA800[H1] that is almost the same)
Not bad at all !!!
Attached Image (click to enlarge)
Click to Enlarge Name: cross ma.jpg Size: 204 KB
  • #5,892
  • Feb 13, 2010 5:40am Feb 13, 2010 5:40am
Quoting maxou888
this is your last suggestion
EMA20[H1] > EMA200[H4] (I put EMA800[H1] that is almost the same)
Not bad at all !!!
Yes, the higher the EMA the safer the trades are.

You need to find a balance between SAFE and the number of potential trade setups.

Scoobs.
  • #5,893
  • Feb 13, 2010 5:42am Feb 13, 2010 5:42am
Scoob,

In my boy I use your idea : Stoch Rsi (M15)
This oscillator gives much more signals than the previous rsi (because of his nature of oscilator) So I think that the trend to take into considération is shorter, so I use if EMA20[M15]>EMA200[M15] && EMA20[H1]>EMA200[H1] that fit better the oscillator frequency (therefore trades frequency)
May be I should add :
if EMA20[M15]>EMA200[M15] && EMA20[H1]>EMA200[H1] && EMA20[H4]>EMA200[H4]
Or obviouly use the stoch RSI on H1 in order to lower the trades frequency
  • #5,894
  • Feb 13, 2010 6:03am Feb 13, 2010 6:03am
Quoting maxou888
Scoob,

In my boy I use your idea : Stoch Rsi (M15)
This oscillator gives much more signals than the previous rsi (because of his nature of oscilator) So I think that the trend to take into considération is shorter, so I use if EMA20[M15]>EMA200[M15] && EMA20[H1]>EMA200[H1] that fit better the oscillator frequency (therefore trades frequency)
May be I should add :
if EMA20[M15]>EMA200[M15] && EMA20[H1]>EMA200[H1] && EMA20[H4]>EMA200[H4]
Or obviouly use the stoch RSI on H1 in order to lower the trades frequency
You should use if EMA20[M15]>EMA200[M15] && EMA20[H1]>EMA200[H1] && EMA20[H4]>EMA200[H4]

M15 = Short term
H01 = Medium term
H04 = Long term

You should always look at 3 different TF's which are at least 4x apart.

However, the list can be endless, so stick with what I have put above, what you need to concentrate on more is the different TP/SL to use per currency traded.

But for the moment stick with the above and see how it works this coming week. There is nothing worse than constantly changing entry conditions as you will end up going around in circles.

Scoobs
  • #5,895
  • Feb 13, 2010 6:26am Feb 13, 2010 6:26am
SCOOBS...
What you think about stop loss < 100
  • #5,896
  • Feb 13, 2010 6:46am Feb 13, 2010 6:46am
Quoting xman
SCOOBS...
What you think about stop loss < 100
It depends on the currency you are trading.

On XXXJPY pairs you may end up blowing allot of trades prematurely.

Commercially I only trade off Daily, Weekly, Monthly, Quarterly charts where the rewards are far higher compared to the initial risk.

But it seems retail traders dont like doing this as they want 100's of trades a week.

Scoobs.
  • #5,897
  • Feb 13, 2010 7:56am Feb 13, 2010 7:56am
I am doing a back testing for only one pair EUR/USD for 01.01.2009-06.04.2009
Results are not a briliant
Attached Image (click to enlarge)
Click to Enlarge Name: StrategyTester.gif Size: 8 KB
  • #5,898
  • Feb 13, 2010 8:02am Feb 13, 2010 8:02am
Quoting xman
I am doing a back testing for only one pair EUR/USD for 01.01.2009-06.04.2009
Results are not a briliant
ST is flawed when it comes to multiple TF's etc.

Steve will confirm this, it is not reliable. You need to forward test.

Also, you only have 25% modelling quality, you need 90% which you wont get because of the flaws in ST.

Scoobs.
  • #5,899
  • Feb 13, 2010 8:17am Feb 13, 2010 8:17am
Yes I undestand. Off course I will put in on the demo from monday.
I just try backtest, in the code I change a Pairs() on one pair EURUSD.
  • #5,900
  • Feb 13, 2010 8:25am Feb 13, 2010 8:25am
Scoobs...
If it is not a secret, can you tell me which broker you are using?
9 traders viewing now