It seems that every time I do research on moving average systems, they tend to perform really well. I am going to forward test in one of my accounts this system I came up with. This may turn out to be a long post but I want the rules to be clear. I am sure that this can be programmed to automation so if someone wants to do backtesting, go for it.
To start, I am going to use the 7 major USD Pairs. Fully invested, each pair will equal 1x my account size. This method is always in the market but the position sizes will constantly be changing.
Here are the EMA crosses to be used:
100/200
50/100
25/50
10/20
5/10
Each cross will be either positive or negative. For example if the 100 is above the 200 it is a (+) and (-) if below. This exercise will be done on all 5 crosses. Each cross will represent 20% of the position. For example if 3 of the crosses are (+) and 2 are (-) the position will be 20% Long 3(20)-2(20)=20. So if the trading account is $10,000, a $2000 position will be taken Long.
If you do the math, you will notice you will always be either long or short 1, 3 or 5 units, long or short.
An always in the market approach is an interesting concept. But as I understand them, MA cross-overs are late/false signals. With 100-200 you don’t get many signals but they will be late/false, with 5-10 cross-overs you will get lots of signals and these will be late/false. There’s no escape - unless you introduce some “moderating” rule?
This is a strange thread… I thought I saw it a few weeks ago… but it is showing as today’s…
Anyway, yes a good system to test, MA Crosses traded with discipline are very profitable.
I used this method when I started, EMA 26 - 200 (read somewhere that this is the settings the insto’s use) and it was the first consistently profitable strategy that worked.
The Bad: When the price was not in a strong trend, the faster EMA (26) tends to bounce above and below the EMA 200 which gave a lot of false signals and with my limited coding skills I couldn’t automate it successfully.
Hence I moved on to a smoothed Heikin Ashi system which removed a lot of noise from the charts.
An EA would have to be able to hold 2 or 3 candles prior to executing a trade in a confirmed direction. A quality programmer could develop this issue out of the EA.
As Tommor stated there can be a lot of time between drinks on the longer EMA’s and absolute roller coasters on the shorter EMA’s but that’s what your research will show you anyway.
Good Luck with your experiment, all those crosses can be profitable with a disciplined approach.
None of it were profitable. I had tested it a year ago before you posted this thread up now on BabyPips. But I do have to say that the 2/8EMA stands out among any other EMA I had seen/heard/used. To believe or not, it’s totally up to you. You can backtest it yourself.
Hey Xavier… its me again…George…George…George of the Jungle… just messin with ya…
Serious stuff…The YouTube video below was the first one that made sense to me as a newbie how to trade an MA Strategy… I have no idea who she is… have no connection to her business (different continents), but after watching this back in late December I started to make some money…
I possibly spent 2 months viewing on Youtube before I found this and I have just supplied the link for FREE and saved you maybe the same amount of time… no…no the pleasure is all yours…
Emulate it EXACTLY… and then tell me with a straight keyboard that an MA Crossover doesn’t work…
In my opinion, as a generalisation, giving figures for the periods of moving average crossovers in the absence of specifying the chart speed is of somewhat limited value, to put it mildly: an EMA 2/8 crossover on an H4 chart, for example, is equivalent to an EMA 8/32 crossover on an H1 chart.
My own perspective of this subject, having ([I]many[/I] years ago) back-tested a very large number of moving average crossovers, on various time-frames, using a decade’s EUR/USD data and some software called Forextester-2, is that in general they can work well in trending markets, but not nearly well enough [I]overall[/I] to be profitable in themselves, i.e. without additional “bias input” from other criteria appropriate to determine whether, according to the parameters needed for whichever is being tested (yes, I know this is a slightly circular argument, and in a sense that’s part of my point), the market is actually “trending” at the time. I also think that in a “clearly trending market” it doesn’t tend to make too much difference exactly which settings are used.
Marcel Link has several points of interest to make on this subject in his well-known book [I]High-Probability Trading[/I], as does F.H. Goslin in his interesting and unusual book [I]Trading Day By Day[/I].
With MA Strategies you have to be creative and think outside the square. I occasionally use another Cross Over Technique where I have the slow EMA 1 and the Fast EMA 100/170/200 (set to move outside of EMA 1 fluctuations).
The fast EMA (1) acts as the price and the slow EMA (~200) as the cross over point on XAUEUR. It can be good for 500 pips per day.
How… My Bot executes a trade on the first (or second) candle to close on the new side of the slow EMA and continues the trade to a preset TP closes the trade then waits dormant until the opposite condition repeats. Effectively a semi auto strategy.
Gold isn’t as “spikey” as currency crosses so on the 1h TF TP is set at ~50 - 70 pips and on the 5m TF ~20 - 50 pips.
Sharper than a normal MA Cross and Play-School simple… just how I like my trading.
I don’t understand why you back-test 10 years… the market cycle changes every month, so I only test over the last 50 days or so emulating “currentish” market conditions. Although I hardly traded this week with the mess due to USD Interest adjustment.
PS. Hey Xavier… I hope you are paying attention… a complete strategy… once again the pleasure is all yours…
True … nothing unusual for me, after all these years, though, with my various problems, etc. I’m generally better than this time last year, though, and much better than a couple of years ago.
(I’ve done only a small handful of trades, this week, but that’s not always a bad thing, in itself. Will start work again on Monday/Tuesday.)
Yes, interesting … and makes sense.
Yes; so I’ve been told, recently … I’ve never actually looked at gold at all … :8:
Neither, with hindsight, do I, really! (I was 17/18 and had bought ForexTester and wanted to play with it, mostly, probably. I’ve also backtested a large number of “systems” from this forum’s “free forex systems” folder :8: ).
I’ve also been very interested, over the last year or so, in seeing how Manxx incorporates moving averages into his trading style, and I have an ex-institutional professional trader friend who now uses MACD’s, apparently very successfully, in her own trading.
Yes, I see … I’ve sometimes looked at “typical price” as an MA setting, which is just (H+L+C)/3, without the open … which I somehow always preferred to just “close”.
