Hanover replies to PMs | Page 17 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Abokwaik replies to PMs 96 replies

Hanover - can you help me with MT4 qs please? 4 replies

Attachments: Hanover replies to PMs
Exit Attachments

Hanover replies to PMs

  • #321
  • Aug 30, 2015 9:31pm Aug 30, 2015 9:31pm
David,
Can you do me a huge favorino and make a slight mod to the plot external data program? I'm hoping you might be able to easily add an "invert data option" to the parameter list.

Please and thank you!
  • #322
  • Aug 30, 2015 10:39pm Aug 30, 2015 10:39pm
Also, is it possible that future data can be displayed on the indicator too? For example, I have a data point for Wednesday, but its only Monday and I want to see that data in the window in the future even though it hasn't passed yet.

Thanks
  • #323
  • Edited 6:14am Aug 31, 2015 4:03am | Edited 6:14am
Quoting brinkclan
Can you do me a huge favorino.......... I'm hoping you might be able to easily add an "invert data option" to the parameter list.
Set the ScalingFactor to -1.

No favorinos for anybody, unless a modification requested is useful to me personally.
Quoting brinkclan
Is it possible that future data can be displayed on the indicator too? For example, I have a data point for Wednesday, but its only Monday and I want to see that data in the window in the future even though it hasn't passed yet.
This is not currently possible. The value displayed under each candle is the data applicable at the candle open. However, you could edit the data in your file to reflect the start date/time of the candle, i.e. in your example, change it from Wednesday's to Monday's date.
____________________

[EDIT] Update to indicator (attached)
As stated above, the value plotted under each candle is the data that applied at the candle open. However, I've added a new parameter PlotDataAtClose, which if set to true, will cause the indy to plot the value that was applicable at each candle close. If data exists for a future candle, then its value will be plotted under the rightmost candle on the chart. E&OE.
Attached File(s)
File Type: ex4 Plot External Data.ex4   26 KB | 347 downloads
  • #324
  • Aug 31, 2015 4:08am Aug 31, 2015 4:08am
thank you. that is sufficient!
  • #325
  • Sep 4, 2015 9:43am Sep 4, 2015 9:43am
Hi

Not strictly a PM but as newbie to ForexFactory I don't seem to be able to send a PM, nor can't I start a new thread.

But in the brief time I've been exploring the forums I've noticed that Hanover is one of, if not the most respected coders on here.

I don't want anyone to do any coding but I'm being frustrated by a problem I've been experiencing and I can't seem to get to the bottom of it.

I want an EA to run so that at the creation of a new bar the EA iterates through 28 pairs and checks the value of a custom indicator for bars [1] & [2] (i.e. not the current bar that has just been created but the previous two bars).

Unfortunately as I iterate through the pairs it seems that some pairs have generated a new candle and some have not.

If I run the EA on GBPUSD and execute the code in the snippet below when a new bar is created on GBPUSD I get the results shown in the screen print.

My question is - if I check bar [1] and bar[2] will I be actually checking bar[1] and bar[2] or will it depend on what iTime is? i.e. will I have to adjust the shift?

Inserted Code
for(int i=0;i<_trading_count;i++)
{
    string symbol=_trading_pairs[i];
    datetime dt=iTime(symbol,Period(),0);
    PrintFormat("%s %s",symbol,TimeToString(dt,TIME_DATE|TIME_SECONDS));
}
Attached Image
  • #326
  • Sep 4, 2015 11:41am Sep 4, 2015 11:41am
Quoting BigOldFella
Hi Not strictly a PM but as newbie to ForexFactory.............
I'm conversant with MQL4 (MT4 build 509 and earlier) but not MQL5 (MT4 build 600 or later). Hence I still do all of my coding using the build 509 compiler, and my reply reflects that. MQL5 may work differently.

There is a function iBarShift that returns the bar# for a given pair/timeframe/bar_open_time. See screenshot. You could use it to ensure that you're working with bars of the same open time across each currency pair.
Attached Image (click to enlarge)
Click to Enlarge Name: PPI 293.png Size: 31 KB
  • #327
  • Sep 7, 2015 3:02am Sep 7, 2015 3:02am
Quoting hanover
{quote} There is a function iBarShift that returns the bar# for a given pair/timeframe/bar_open_time.
Brilliant, thanks. I'll give that a try.
  • #328
  • Edited 7:48pm Sep 16, 2015 6:13pm | Edited 7:48pm
Quoting hanover
I'm conversant with MQL4 (MT4 build 509 and earlier) but not MQL5 (MT4 build 600 or later).
Hi David - do you know of any coders for hire whom you'd recommend? (Feel free to pm me if you'd rather not talk about this in public)

I'm looking for someone to code a simple ea (I know all non-coders say it's "simple", but the ea-coding husband said it shouldn't be more than a 2-hr job for a professional) using three currently available indicators (which need no mods). I'm really just looking for an ea for backtesting purposes.

I know I could trawl through elance etc but I prefer to go via recommendation, and I trust you to be a good judge of coders
  • #329
  • Oct 6, 2015 11:28am Oct 6, 2015 11:28am
Hello David!

I'd like to add a simple function to one of your best (for me the best) indicators: Strengh Indicator - Giraia.
Could you please tell me if you can do it or refer someone? PM me.

Thanks a lot for fine works you have been spreading around this forum.

Caatinga
  • #330
  • Oct 7, 2015 1:18am Oct 7, 2015 1:18am
Quoting Caatinga
Hello David! I'd like to add a simple function to one of your best (for me the best) indicators: Strengh Indicator - Giraia. Could you please tell me if you can do it or refer someone? PM me. Thanks a lot for fine works you have been spreading around this forum. Caatinga
Hi,

I regard 'Currency strength - Giraia' as the least useful of the five strength meters that I've written. I guess one person's trash is another's treasure.

I get a lot of programming requests, and I just don't have the time to look at them all. Hence nowadays I undertake only those that I think might be useful to me personally. I apologize for any inconvenience.

I suggest posting your request in one of the 'I code your indicators for free' threads. Apparently there are now multiple versions of the indy floating around, so please be sure to attach the .mq4 file that you want upgraded with your enhancement. The indy was originally written a few years ago using the build 509 compiler, so it will either first need to be updated to run compatibly on the latest builds, or the changes will need to be made according to this post. Also, if the version you want updated calls some of the 'hanover --- extensible functions' routines, then you will need the files either here (build 509) or here (build 600+). Don't worry if all of this seems complex, a MT4/MT5 programmer will understand what I'm talking about; just refer him to this post. Beyond that I'm happy to answer any questions he might have.

If you don't get a response, then please feel welcome to post again, and I will PM you with the names of some very capable (albeit commercial) MT4 programmers.

Good luck,
David
  • #331
  • Oct 7, 2015 1:31am Oct 7, 2015 1:31am
Quoting hanover
{quote} Hi, I regard 'Currency strength - Giraia' as the least useful of the five strength meters that I've written. I guess one person's trash is another's treasure. I get a lot of programming requests, and I just don't have the time to look at them all. Hence nowadays I undertake only those that I think might be useful to me personally. I apologize for any inconvenience. I suggest posting your request in one of the 'I code your indicators for free' threads. Apparently there are now multiple versions of the indy floating around, so please be...
Hey !

Thanks for your kind answer!
Yes, you are right, what is not good for some is good for others...please PM me and I'll show you that in some nice graphical manner...
I'd really like to have the names of those "very capable (albeit commercial) MT4 programmers", please.

Thank you.
  • #332
  • Oct 7, 2015 3:01am Oct 7, 2015 3:01am
Hi

Quoting hanover
{quote} Hi, I regard 'Currency strength - Giraia' as the least useful of the five strength meters that I've written.
I think you might be able to guess my next question.

If Giraia is the least useful, what is the most useful? Or do they all have different attributes?
  • #333
  • Edited 9:32am Oct 7, 2015 5:19am | Edited 9:32am
Quoting BigOldFella
If Giraia is the least useful, what is the most useful? Or do they all have different attributes?
BOF,

The 5 indicators are Recent Strength (RS), CSM, J4D, PSA and Giraia.

If I was pressed into picking a favorite, it would be RS. It's not because it uses a more accurate or relevant calculation. Instead, to me, as a discretionary trader, the lines (representing continuous values) paint a more comprehensive picture than the static values in CSM's histogram. (The latter would perhaps be more suited to an automated trading system, where a single, discrete value is more practical and concise). Also, the RS has a wider range of parameters, allowing greater customization. Both RS and CSM employ different calculations, and hence I find CSM useful to confirm what RS is showing me. For what it's worth, CSM is more lightweight (uses less computer CPU/RAM) and installs much more easily.

J4D was a specialized offshoot of CSM that I wrote for another FF member. I don't use it at all myself.

PSA plots discrete values for both pairs and currencies, across multiple timeframes with different weightings per timeframe, and is also extremely configurable. It uses a third type of calculation, namely averaging the slopes of regression lines. When I first wrote it, I used it a lot, but its novelty has worn off somewhat, and nowadays I prefer RS and CSM. But again, due to its very different calculation method, it can provide quasi-independent confirmation.

Giraia is an indicator that was modelled at the request of FF member giraia_br, and uses the calculations from a XLS that he provided. I dislike it for two main reasons: (1) the way it rescales its calculation into bands, and then assigns a value for the band regardless of where the actual value was within the band; and (2) it resets itself at MT4 midnight every day. This means that the values are too sensitive to be accurate as the beginning of a new day, when the values swing wildly and are basically meaningless. Where I live (New Zealand) MT4 midnight occurs 9am-11am NZT (depending on daylight saving), and the values don't stabilize until much later in the day, making the indy useless for trading a large part of the Asian session.

I've always believed is that it's not the indy that's important, but the way that the trader interprets the values (which can differ greatly, of course, depending on how the values are calculated), and incorporates it into his total trading methodology. Hence (apart from the limitations I mentioned with Giraia) I don't believe that any one indicator's calculation is superior to another's. For me, trading is an art rather than a science, and I believe that folk get hung up on the accuracy of numbers, whereas my approach is more broadbrush (and hence I don't find threads like this one to be overly helpful). Markets are driven by the volumes of orders that exist at different levels, which -- especially given the paucity of data that is available to the retail trader -- can not be nailed down in a precise mathematical formula. Moreover, I believe there's a case for not using a strength meter indicator at all, as web pages like this and this are perfectly adequate in defining currency strength and weakness, IMO.

David

Quoting Caatinga
I'd really like to have the names of those "very capable (albeit commercial) MT4 programmers", please.
I will search thru my e-mails for the list that I sent to another FF member, and PM you tomorrow.
  • #334
  • Oct 7, 2015 6:13am Oct 7, 2015 6:13am
Quoting hanover
{quote}I will search thru my e-mails for the list that I sent to another FF member, and PM you tomorrow.
Thanks!
  • #335
  • Oct 7, 2015 9:38am Oct 7, 2015 9:38am
@bugsbunny110, robboyes, theprofessor: sorry, but I'm not coding EAs for traders any more, unless the trader can show an objectively verifiable record of profitability for at least 3 years. And even then I would probably need more info about the system, especially how it manages risk. More than 99% of EAs are unprofitable; no EA can trade as intelligently and adaptively as an experienced human trader.

@Saronko: thanks for your kind words.
  • #336
  • Oct 7, 2015 10:39am Oct 7, 2015 10:39am
Quoting hanover
{quote} BOF, The 5 indicators are Recent Strength (RS), CSM, J4D, PSA and Giraia.
Thank you once again.

I shall have a look at each.
  • #337
  • Oct 8, 2015 1:55am Oct 8, 2015 1:55am
Hanover, great of you to answer questions in this thread, maybe you can answer mine? I know no one better suited than you, to ask this question to, I have a problem with Mt4 that makes me crazy, and I don't know what is the reason, the problem is that I am getting ordersend error 131 on all the experts I am trying to backtest, all the EA's refuse to open trades, even MT4s own sample macd and ma experts, there can't be any fault on the sample EA's, can it? I have no clue what to do, the history seems not to be the problem either, I have downloaded it many times from metaquotes, I know it is probably not the best quality but it can't be that bad? apprichiate any help.​
Bulls are stupid Animals!especially when Im short!
  • #338
  • Oct 8, 2015 3:21am Oct 8, 2015 3:21am
Quoting roughtrader
Hanover, great of you to answer questions in this thread, maybe you can answer mine? I know no one better suited than you, to ask this question to, I have a problem with Mt4 that makes me crazy, and I don't know what is the reason, the problem is that I am getting ordersend error 131 on all the experts I am trying to backtest, all the EA's refuse to open trades, even MT4s own sample macd and ma experts, there can't be any fault on the sample EA's, can it? I have no clue what to do, the history seems not to be the problem either, I have downloaded...
I hope I'm not being rude to Hanover but I think I can answer this one. It's to do with lot sizes. They have to be multiples of a certain value.

You can find this value out by using
Inserted Code
MarketInfo(symbol, MODE_LOTSTEP)

So you need something like this to round your calculated lot size down to the nearest lot step: -

Inserted Code
double lot_step = MarketInfo(symbol, MODE_LOTSTEP);
lot_size = MathFloor(lot_size/lot_step) * lot_step;


If you want to round up or down depending on which is the nearest use: -
Inserted Code
double lot_step = MarketInfo(symbol, MODE_LOTSTEP);
lot_size = MathFloor(lot_size/lot_step + 0.5) * lot_step;

Let me know if it works.
  • #339
  • Edited 4:41am Oct 8, 2015 4:21am | Edited 4:41am
Quoting roughtrader
Hanover, great of you to answer questions in this thread, maybe you can answer mine? I know no one better suited than you, to ask this question to, I have a problem with Mt4 that makes me crazy, and I don't know what is the reason, the problem is that I am getting ordersend error 131 on all the experts I am trying to backtest, all the EA's refuse to open trades, even MT4s own sample macd and ma experts, there can't be any fault on the sample EA's, can it? I have no clue what to do, the history seems not to be the problem either, I have downloaded...
Rough,

Sorry, I don't know the answer to this one. I would expect the factory supplied EAs to calculate the lot size correctly.

BigOldFella is right. Error 131 is normally caused by a calculated lot size being either (1) less than the minimum allowed lot size (with most br0kers, that's 0.01), (2) greater than the maximum allowed lot size (with most br0kers, that's 100.00), or (3) not a multiple of the br0ker's lot step (for example, if this is 0.01, then the lot size must be 0.01, 0.02, 0.03, etc, e.g. 0.025 would cause the error). If the EA calculates the lot size as X% of capital, the remaining account balance is small and the SL is wide, then the lot size may be < 0.01, or possibly even rounded down to 0, which would cause error 131. That's the most common reason I've encountered, that gives error 131. So one thing to try would be to open a new demo account with a balance of say $100,000.

Some other obvious things to ensure are to make sure that 'Allow automated trading' is checked on, and also 'Allow live trading'. And also that the 'Auto Trading' button on MT4's toolbat is on (green), which should result in a smiley face showing next to the EA name. But failure to do this should cause a different error than 131.

Sorry, but beyond that I'm stuck. I'm running Pepperstone MT4 build 745 and (coincidentally) ran a backtest using the factory supplied MACD EA earlier this evening (was testing to see if a certain indy was repainting), and it placed trades quite happily. Perhaps there is a bug in builds later than 745, I don't know. You could also try downloading a different br0ker's MT4 and see if the same problem occurs there. Or simply contact your br0ker's helpdesk: since you're paying them transaction costs per trade, they should be eager to find a solution for you if they want to retain your business.

David

[EDIT] Here is the standard piece of code that I use to ensure that the lot size is correctly normalized (it's similar to BigOldFella's, although I round everything down):

Inserted Code
  lotstep  = MarketInfo(ccy,MODE_LOTSTEP);
 
........
 
  if (lotstep > 0)
    vol = lotstep * MathInt(vol/lotstep);
  if (lotstep == 0.01)   vol = NormalizeDouble(MathInt(vol,2),2);
  if (lotstep == 0.1)    vol = NormalizeDouble(MathInt(vol,1),1);
  if (lotstep == 1)      vol = NormalizeDouble(MathInt(vol,0),0);
 
.......
 
//+------------------------------------------------------------------+
double MathInt(double n, int d=0)  {
//+------------------------------------------------------------------+
// Corrects a rounding/accuracy bug in MQL4's MathFloor function
//   (use MathInt(n) instead of MathFloor(n)
// Rounds n DOWN to d decimal places, e.g.
// MathInt(2.57,1) returns 2.5
// MathInt(2.99)   returns 2
   return(MathFloor(n*MathPow(10,d)+0.000000000001)/MathPow(10,d));
}
_____________________

Quoting BigOldFella
I hope I'm not being rude to Hanover
Hell no, I'm always delighted if somebody can save me some typing by solving problems directed at me. So, many thanks, I owe you a beer!
  • #340
  • Oct 8, 2015 4:31am Oct 8, 2015 4:31am
hello hanover ,

may i spent your time please , this indy didn work, can you help.
Attached File(s)
File Type: mq4 Elliott_Waves.mq4   27 KB | 511 downloads


thanks
0 traders viewing now