Why exactly do EMA crossover systems not work? | Page 9 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Help. EMA Crossover with different EMA cross for exit 0 replies

EMA crossover (ema 14 & 34) 4 replies

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

Systems, Systems and Systems... 1 reply

EMA Crossover Systems 1 reply

Attachments: Why exactly do EMA crossover systems not work?
Exit Attachments

Why exactly do EMA crossover systems not work?

  • #161
  • Jul 14, 2011 5:14am Jul 14, 2011 5:14am
Quoting nubcake
yup.

i should just not bother arguing with people who simply won't know any better. i should just jump in the eu thread and make random calls for amusement instead and see how many sheep i can herd.

The issue is, your arguing with people and sticking to your guns / belief which is fine, but some times like yesterday your just plain wrong end of story, sorry but that's the case.

EU thread, seems to be much more intelligent than GU, so go do that on GU!!

Post 5 predictions, then big the 1 up that works out later, hours of fun.
Nothing to it, but to do it!!! Stick to the plan FOOL!!!!
  • #162
  • Jul 14, 2011 5:25am Jul 14, 2011 5:25am
Quoting Turveyd
The issue is, your arguing with people and sticking to your guns / belief which is fine, but some times like yesterday your just plain wrong end of story, sorry but that's the case.
yesterday was a long time ago for me... plz remind me where i was wrong?
Congratulations ScalaFX for coming out of the closet!
  • #163
  • Jul 14, 2011 5:29am Jul 14, 2011 5:29am
Quoting nubcake
yesterday was a long time ago for me... plz remind me where i was wrong?

Likewise, just trust me your human you get things wrong as do we all, drop the superior attitude and remember 2 years in your still just a demo trader
Nothing to it, but to do it!!! Stick to the plan FOOL!!!!
  • #164
  • Jul 16, 2011 7:27am Jul 16, 2011 7:27am
Quoting Turveyd
Nothing I have said in the past few years is worth anything. I now realise, after doing nothing but losing money for years, that there is no way I should be giving advice to others. The truth is, I just can't trade.
We knew that all along.
  • #165
  • Jul 16, 2011 1:32pm Jul 16, 2011 1:32pm
Quoting TheGreatMilenko
We knew that all along.
just.... wow. your post history, especially the extracted stuff that sadly no longer appears in the original threads... wow.

i don't care if you know me or not or think i am a retard or a fool, but I LIKE YOU! if only i had known about all of your posts a few years ago that pretty-much cover everything i have been saying to turveyd recently i would have just linked to your profile and not wasted my time!

you've already said it all before, as has he.
Congratulations ScalaFX for coming out of the closet!
  • #166
  • Jul 16, 2011 1:37pm Jul 16, 2011 1:37pm
Quoting nubcake
just.... wow. your post history, especially the extracted stuff that sadly no longer appears in the original threads... wow.

i don't care if you know me or not or think i am a retard or a fool, but I LIKE YOU! if only i had known about all of your posts a few years ago that pretty-much cover everything i have been saying to turveyd recently i would have just linked to your profile and not wasted my time!

you've already said it all before, as has he.
That confirms it willing to side with the biggest loser on the site you really are a total A HOLE!

Not profitable demo only pro trader lol
Nothing to it, but to do it!!! Stick to the plan FOOL!!!!
  • #167
  • Jul 16, 2011 1:50pm Jul 16, 2011 1:50pm
the title of this thread says EMA. not SMA. not MA. would someone really think that the extra weighting in an EMA calculation would be all the difference required to go from random unprofitable trades to logical profitable trades? is that the thinking some have? i don't have a great memory for unimportant details and don't recall my own initial thoughts all those moons ago, so i'm curious if anyone remembers ever thinking along these lines at all?
Congratulations ScalaFX for coming out of the closet!
  • #168
  • Jul 16, 2011 2:06pm Jul 16, 2011 2:06pm
Quoting nubcake
the title of this thread says EMA. not SMA. not MA. would someone really think that the extra weighting in an EMA calculation would be all the difference required to go from random unprofitable trades to logical profitable trades? is that the thinking some have? i don't have a great memory for unimportant details and don't recall my own initial thoughts all those moons ago, so i'm curious if anyone remembers ever thinking along these lines at all?
There are MAs and there are MAs.

Here's a "quant EMA":

http://dl.dropbox.com/u/190212/ff/quant_ema_1.png

http://dl.dropbox.com/u/190212/ff/quant_ema_2.png

http://dl.dropbox.com/u/190212/ff/quant_ema_3.png

http://dl.dropbox.com/u/190212/ff/quant_ema_4.png

http://dl.dropbox.com/u/190212/ff/quant_ema_5.png
  • #169
  • Jul 16, 2011 3:01pm Jul 16, 2011 3:01pm
Quoting Adal
There are MAs and there are MAs.

Here's a "quant EMA":
that's gnarly. i will have to read again after a sleep. what could one glean from such ma's? if i'm seeing it in my head correctly then it possibly provides an algorithmic way to roughly determine values for a certain periods range/volatility depending on the weighting? i think i'm way off target.

quants. my kryptonite.
Congratulations ScalaFX for coming out of the closet!
  • #170
  • Jul 16, 2011 7:18pm Jul 16, 2011 7:18pm
Quoting nubcake
would someone really think that the extra weighting in an EMA calculation would be all the difference required to go from random unprofitable trades to logical profitable trades? is that the thinking some have?
Sadly enough, a lot of people do, even the "smart" ones.

There is a big difference between WMA, SMA, EMA, and all other newer _MA. It's 100% correct that the newer _MA being produced are able to catch an earlier move or better curve. What they don't realize is that the more you curve a MA, the higher it produces false signals during consolidation periods. Thus, rendering it's efficiency if you don't know how to use it.

This is no different from car manufacturers trying to outperform Porsche. You see, they come out with higher kilowatts, higher torque and more power, more than a Porsche.

Yet Porsche would outperform them all with a simple philosophy:
"There is no point in increasing power if you can NOT control it. Uncontrolled power is lost performance and also decreases performance efficiency greatly."

Porsche outperform its competitors because the car has rock-solid handling. And this handling comes from the balance or controlling it's engine power... and not because of having more power.

So I'm going to make it clear right here:
Improving or enhancing MA is a F*cking waste of time if you don't know how to use it. If you are not profitable in using MA, then EMA, WMA, SMMA, LWMA, etc... will never make you any profitable.
  • #171
  • Jul 16, 2011 7:37pm Jul 16, 2011 7:37pm
Quoting CrucialPoint
Sadly enough, a lot of people do, even the "smart" ones.

There is a big difference between WMA, SMA, EMA, and all other newer _MA. It's 100% correct that the newer _MA being produced are able to catch an earlier move or better curve. What they don't realize is that the more you curve a MA, the higher it produces false signals during consolidation periods. Thus, rendering it's efficiency if you don't know how to use it.

This is no different from car manufacturers trying to outperform...

Fund / big banks buying systems are based around EMA's mainly it would seem, not all, they vary but on average enough to make some MA's of different levels work.

Not talking about EA's here, just simple bits of logic that place the order for 10,000 lots and accumulate in the background for whoever tells it to do it, thats what gives some MA's there power.

MA covers all the different MA's, so EMA, SMA, LWMA and what ever.

MACD is just the difference between 2 MA's. most indicators are based around MA's of a kind.
Nothing to it, but to do it!!! Stick to the plan FOOL!!!!
  • #172
  • Jul 16, 2011 7:46pm Jul 16, 2011 7:46pm
Quoting Turveyd
Fund / big banks buying systems are based around EMA's mainly it would seem, not all, they vary but on average enough to make some MA's of different levels work.

Not talking about EA's here, just simple bits of logic that place the order for 10,000 lots and accumulate in the background for whoever tells it to do it, thats what gives some MA's there power.
And that's the type of mentally that caused the financial collapse... because they thought they have it under control.
They took more risk believing that they can predict the future or believing in their power to make more money that they forgot to look on the downsides and the risks.
  • #173
  • Jul 16, 2011 7:47pm Jul 16, 2011 7:47pm
IMHO, on ma's:
It's beyond me why anyone would average a set number of bars in the next (ma indicator's) bar, when the chart's character can dramatically change due to what's within each bar.
  • #174
  • Jul 16, 2011 7:51pm Jul 16, 2011 7:51pm
Quoting nubcake
that's gnarly. i will have to read again after a sleep. what could one glean from such ma's? if i'm seeing it in my head correctly then it possibly provides an algorithmic way to roughly determine values for a certain periods range/volatility depending on the weighting? i think i'm way off target.

quants. my kryptonite.
You just use it instead of a normal EMA. They measured the profitability of standard EMA versus iterated EMA in a series of simple trading systems, including a moving average crossover system. The iterated EMA was 4% more profitable then the standard EMA. Both were unprofitable if you took transactions costs into account, but the "quant EMA" was closer to break-even.

This is how quants work. They try things like all of us, but they MEASURE them. And they take each improvement and incorporate it into their framework. Then they look again for things. This way, each year the equations are improved and they more accurately predict things (look at GARCH evolution for example). And sometimes, they discover brand new things (like the ACD model).

Meanwhile, retail traders keep using the same old 50-year old simple MA. No improvement on this side of the fence It's like progress stopped. The 50 year old car is good enough, it's about how you use it, right. We don't need to stinking ABS, traction control or other improvements.
  • #175
  • Jul 16, 2011 7:52pm Jul 16, 2011 7:52pm
Quoting ha-pattern
IMHO, on ma's:
It's beyond me why anyone would average a set number of bars in the next (ma indicator's) bar, when the chart's character can dramatically change due to what's within each bar.
To minimize and filter noise
  • #176
  • Jul 16, 2011 7:57pm Jul 16, 2011 7:57pm
Quoting Adal
Meanwhile, retail traders keep using the same old 50-year old simple MA. No improvement on this side of the fence It's like progress stopped. The 50 year old car is good enough, it's about how you use it, right. We don't need to stinking ABS, traction control or other improvements.
And this is the type of mentally why NASA invested $1,000,000 in inventing a pen that works in zero gravity.... The Russians used an ancient technology that didn't cost a thing...

They used a pencil.
  • #177
  • Jul 16, 2011 7:58pm Jul 16, 2011 7:58pm
here is a screen shot I posted on my other thread, I trade entirely of MA's and Envelopes. ( EU M15 )

The twin red lines was 250ema, I don't use 0 pip width I used a envelope to allow for any inaccuracy.

See how many times the market moved upto that level, selling picked up and down she went again, that is how I use EMA's.

Most of the time if it turns over then it keeps going up, obviously if you shorted below when that happens you'll take a small loss, but there are 5 short ops over nearly 4 days there without 1 loss!!

It doesn't always look this good ofcourse, but it does highlight others are using it and it's definately an edge the more edges you can get the better.
Attached Image (click to enlarge)
Click to Enlarge Name: System better explained.jpg Size: 457 KB
Nothing to it, but to do it!!! Stick to the plan FOOL!!!!
  • #178
  • Jul 16, 2011 8:00pm Jul 16, 2011 8:00pm
Quoting CrucialPoint
And this is the type of mentally why NASA invested $1,000,000 in inventing a pen that works in zero gravity.... The Russians used an ancient technology that didn't cost a thing...

They used a pencil.

Although it's a good example kinda, it's actually not true the Americans used pencils to, maybe they couldn't get the pen to work despite spending millions, just an old myth.

The old pressurised pen that works upside down on QVC LOL
Nothing to it, but to do it!!! Stick to the plan FOOL!!!!
  • #179
  • Jul 16, 2011 8:02pm Jul 16, 2011 8:02pm
Quoting ha-pattern
IMHO, on ma's:
It's beyond me why anyone would average a set number of bars in the next (ma indicator's) bar, when the chart's character can dramatically change due to what's within each bar.
Many of the current models work on ticks. So even if it was something slow changing like a "daily" moving average, it's actually recomputed each tick and also for each tick in the past.

MetaTrader for example does recompute the moving average on each incoming tick, but for past values it uses bars, which loses information like you said.

CrucialPoint, you really should be checking your sources. Because otherwise you are just repeating the same nonsense that you read. About the NASA pen for example: http://www.snopes.com/business/genius/spacepen.asp I like to have references for everything I say. Instead you just declare that "all moving averages are equal" and the only thing it matters is "how you use them".
  • #180
  • Edited 8:57pm Jul 16, 2011 8:27pm | Edited 8:57pm
Quoting Adal
Many of the current models work on ticks. So even if it was something slow changing like a "daily" moving average, it's actually recomputed each tick and also for each tick in the past.

MetaTrader for example does recompute the moving average on each incoming tick, but for past values it uses bars, which loses information like you said.
Sorry, no, you miss my point (which Crucialpoint understood well enough.)
Another way to say it:
Each bar has a distinct position in relation to every other bar, such that groups of bars (which one reason or another defines) follow pivotal bars that all align to create the next chart move.
If you simply treat every bar -- tick or otherwise -- like the last one serially, then the chart study ignores position.

One can argue there is an infinite number of groups of bars to which a bar can have the definition of its position changed, and thus one can treat the entire price-time line as having bumps a ma or similar is selected/trained to read somewhat well.

I prefer to believe there is hope to discern which groups of bars most define the next bar's position.

Honors to your knowledge, though, and its equally valid philosophy toward charts.

I don't know, maybe ma's can work for very short periods, I have seen some good things from such indicators, such as parallel ma's serve as channels a little better on occasion. It's just the underlying makeup of ma's that disturbs me, how it ignores news, order flow / spread / etc., bar patterns, all of this (and all reasons to make groups of individually-picked bars), for the sake of uniformity.
0 traders viewing now