MACD / Stochastic Combo Strategy 4-Hour

platinyumnum

Junior member
Messages
15
Likes
2
MACD / Stochastic Combo Strategy 4-Hour.

This posting is for strategy setting up and testing rather than for following in exact detail as listed below. I have been getting six figure results for about a year and a month of Dukascopy historical data.

EUR/USD 4hr Dukascopy
Dukascopy 4hr history created by taking the 1hr history and getting each 4 hours worth of high highs and low lows from it.

Dukascopy - only use the 4hr periods that have their hours listed as 0, 4, 6, 12, 16, 20. The periods listed as 1, 2, 3, 5, 6, 7, 9, 10, 11, 13, 14, 15, 17, 18, 19, 21, 22, 23, probably won't work well.

Dukascopy - Delete all the hours that don't have any volume because the system was closed for weekends or holidays. Delete them so that their empty calc won't get in the way of calculations (for those of you doing Excel calculations). If you are using some form of charting where you can't add your own figures, you may have to accept what you get. This works almost as well.

MACD = 6/36/14.
Fast Stochastic = 8/6.
BASICALLY - What we are doing is transacting according to the MACD with guidance from the Stochastic.

Stop = not less than 277.
Limit = Not more than 366.
Transact no more than 1 contract (NOT ONE LOT OF 10 CONTRACTS) per every $400 of equity in the account - at 200 to 1.

Basically...
Buy when the MACD moves above its signal.
Sell when the MACD moves below its signal.

If the difference between the Stochastic and its signal is less than 10, ignore Opening a trade.
If the difference between the price this close and the price last close, divided by the price last close, times 10,000, is greater than 10, ignore Opening a new trade even if the other signals say to do it. Formula: (Absolute Value ($$1 - $$2) / $$2 x 10,000) > 10.
If the MACD indicates that you should open a trade, but the Stochastic or the price keep you from doing so, if the Stochastic or price free you up several periods later, go ahead and Open the trade provided the MACD is still in the same signal position.

If you get a signal to Buy or Sell, and you have an opposite trade open already, from a previous signal, close the old trade, even if it is a loss, and open the new opposite one.
Leave every trade open until it Stops or Limits, or is shut out by an opposite trade signal.

THIS SOUNDS COMPLICATED, BUT...

Set up a chart to the settings and take a look. The charts I tested for this were FxTrek, Netdania and WinTrader.

If you do your own testing in Excel, be careful. It's easy to mess things up, get good signals, and yet be completely wrong!

So far, after a week or so of detailed testing on live charts, with real figures, watching real time every 4 hours, it is working up to the historical tests specs. No live money testing done yet.

You know how the Forex goes. Test continually.

Will it work with other currency pairs? Don't know.

I make my share of mistakes, so if somebody finds that this won't work, post. Thanks.

:)
 
MACD / Stochastic Combo Strategy 4-Hour.

This posting is for strategy setting up and testing rather than for following in exact detail as listed below.

EUR/USD 4hr Dukascopy

Dukascopy 4hr history created by taking the 1hr history and getting each 4 hours worth of high highs and low lows from it.

If you are using some form of charting where you can't add your own figures, you may have to accept what you get. This works almost as well.

MACD = 6/36/14.
Fast Stochastic = 8/6.

THIS SOUNDS COMPLICATED, BUT...

Set up a chart to the settings and take a look.
The charts I tested for this were FxTrek, Netdania and WinTrader.

So far, after a week or so of detailed testing on live charts, with real figures, watching real time every 4 hours, it is working up to the historical tests specs. No live money testing done yet.

Will it work with other currency pairs? Don't know.

I make my share of mistakes, so if somebody finds that this won't work, post. Thanks.

:)
Hi Platinyumnum

Sounds like your rationale is good. Can you please post a screen shot of your chart(s)?

Unfortunately I do not have EXCEL skills to be able to filter data as you advise.

How do your signals compare when using unfiltered data?
Are the results very different using unfiltered data?
Can the method be adapted in any way to suit the MT4 platforms?

If you think it wothwhile I will have a go using MT4 data, but I am not au fait with back-testing either, so will probably have to rely on your tweaks.

Hats off to you for this innovative approach. I look forward to following along - even if I can not reproduce in MT4.

Kind regards

Ingot.
 
Hi Platinyumnum

Sounds like your rationale is good. Can you please post a screen shot of your chart(s)?

Unfortunately I do not have EXCEL skills to be able to filter data as you advise.

How do your signals compare when using unfiltered data?
Are the results very different using unfiltered data?
Can the method be adapted in any way to suit the MT4 platforms?

If you think it wothwhile I will have a go using MT4 data, but I am not au fait with back-testing either, so will probably have to rely on your tweaks.

Hats off to you for this innovative approach. I look forward to following along - even if I can not reproduce in MT4.

Kind regards

Ingot.

Thanks for your reply.

Two of the charts are free, online. You can set them up yourself, easily.
The Netdania chart is: Live forex charts, real-time forex charts, live currency charts .
The FxTrek Intellichart is: Foreign Exchange Charts Panel .
The third chart, WinTrader, offers no advantage offer the other two.

I'm not sure what you mean by "filtered" data. EXCEL simply makes it easy to "manipulate" lots of data with mathematical functions. One can design the mathematical functions of the MACD and the Stochastic in EXCEL, and then apply them to the Forex historical data within controlled parameters that one designs.

Raw data doesn't give any results except what you see - the price goes up or the price goes down, and once in awhile it doesn't change. Is that what you mean by "unfiltered" data?

When one looks at a chart, one sees a picture of what is happening, sort of like an overview of the mathematical manipulation of the data. The behind-the-scenes-mathematical-manipulation produces quite exact numbers, that are usually not accurately portrayed in a chart view. If one is not familiar with mathematical formulations, then the chart might be the only option - along with the "guesses" that go with chart views.

Thanks again.

:)
 
Thanks for your reply.

Two of the charts are free, online. You can set them up yourself, easily.
The Netdania chart is: Live forex charts, real-time forex charts, live currency charts .
The FxTrek Intellichart is: Foreign Exchange Charts Panel .
The third chart, WinTrader, offers no advantage offer the other two.

I'm not sure what you mean by "filtered" data. EXCEL simply makes it easy to "manipulate" lots of data with mathematical functions. One can design the mathematical functions of the MACD and the Stochastic in EXCEL, and then apply them to the Forex historical data within controlled parameters that one designs.

Raw data doesn't give any results except what you see - the price goes up or the price goes down, and once in awhile it doesn't change. Is that what you mean by "unfiltered" data?

When one looks at a chart, one sees a picture of what is happening, sort of like an overview of the mathematical manipulation of the data. The behind-the-scenes-mathematical-manipulation produces quite exact numbers, that are usually not accurately portrayed in a chart view. If one is not familiar with mathematical formulations, then the chart might be the only option - along with the "guesses" that go with chart views.

Thanks again.

:)
Thanks Plat

I'll check the charting links.

What I meant by "filtered" data was the exclusion of certain data as per your "Delete all the hours that don't have any volume" rule. That is quite an innovative approach, and one that would retain only data that has the weight of money behind it - ie a conviction about price by the participants.

Not only does this '"refined" data alter the shape of a chart, it also alters the signal given from an indicator derived from that data ... a Win : Win situation for the trader.

I had a look at the two charting sites - and for a MT4 user, it is a bit of a learning curve to get the head around. I can not find where to delete the low volume bars on either of these charts. I assume you need to be using EXCEL to manipulate the data in the way you describe. Any suggestions? Also the construction of a fast Stochastic 8/6 is a bit elusive. Can you be more specific? I would like to at least draw these charts in MT4 and even if I can not delete the low volume bars, I may be able to find signals according to your other criteria.

I like your ideas Plat - keep them coming. Pity I didn't have some Excel skills -I'd be all over this like a rash.

Best wishes

Ivan
 
Thanks Plat

I'll check the charting links.

What I meant by "filtered" data was the exclusion of certain data as per your "Delete all the hours that don't have any volume" rule. That is quite an innovative approach, and one that would retain only data that has the weight of money behind it - ie a conviction about price by the participants.

Not only does this '"refined" data alter the shape of a chart, it also alters the signal given from an indicator derived from that data ... a Win : Win situation for the trader.

I had a look at the two charting sites - and for a MT4 user, it is a bit of a learning curve to get the head around. I can not find where to delete the low volume bars on either of these charts. I assume you need to be using EXCEL to manipulate the data in the way you describe. Any suggestions? Also the construction of a fast Stochastic 8/6 is a bit elusive. Can you be more specific? I would like to at least draw these charts in MT4 and even if I can not delete the low volume bars, I may be able to find signals according to your other criteria.

I like your ideas Plat - keep them coming. Pity I didn't have some Excel skills -I'd be all over this like a rash.

Best wishes

Ivan

Thank you.

The historical data I use for semi-long-term Excel testing comes free from Dukascopy http://www.dukascopy.com/swiss/english/data_feed/csv_data_export/ . Dukascopy historical data includes virtually all hours - including weekends and holidays when the Forex is shut down. Since the Forex is not operating at these times, no volume is generated (except by a few small groups on rare occasion). Yet Dukascopy includes these times in their data as "zero-volume" periods. Since they interrupt the continuity of the indicators anybody uses, I take them out of the data to get a smooth flow.

MT4 is a better charting concept in my opinion - at least for user manipulation of the indicators. But I think I read in their literature somewhere that they have slightly different formulas for some of their indicators. Priority formulas that they figured out themselves. Their MACD seems to be some kind of combination of the two MACD indicators in the FxTrek chart. But I'm not sure. I like MT4.

All charting systems are at least slightly different.

The few online charting systems that I have worked with seem to delete, automatically, weekend zero-volume periods.

MT4 Stochastic is 8/1/6 as is the one for Netdania.

The whole concept behind the use of indicators is the "deleting of," or "limiting of," certain bars according to the parameters set up by the indicators. And you are the one who sets up the indicators. So, you control the show.

My system is for those who want to use detailed, non-charting, crunch-the-numbers type of set-ups, for getting a more exact idea of what happened than the charts can give. Like Excel, or other spreadsheet programs.

Thanks.
 
Top