Damnit! Lost a whopping $13.5 today on XAU. Keep up with this downlegg XAU; I'll just buy more of you. Long 1726.22 Long 1742.84 Long 1733.60 Target: Above 1800 by end of next week. Stop: 1650 Average purchase price: 1734.22. Closing price Feb 08: 1732.72 Down: 1.5 X 9 = $13.5. So what's your truth, GrandSuperPoop?
It's very easy to find an indicator look back period combo that would have worked in hind sight. It doesn't mean it will continue to do work. I also noticed several false signals that you did not annotate on your chart. You'll want to factor those losses into your assessment.
Also those moving averages look different after the fact than they do at the instant you make your trade in real time. Realize that the moving average recalculates as the prices come in. As a result the position of the line will change. Therefore, what looks like a cross now, may not have been in real time.
No need to "cut hairs". You referenced an entry by N54_fan. The anotation comment is very correct. http://www.elitetrader.com/vb/showthread.php?s=&threadid=214546&perpage=6&pagenumber=269
As I didn't annotate any entry or exit signal in the aforementionned chart but quoted the poster to adress him, I don't think it's correct at all. But might be an english language issue. Also the question I asked, is if anyone ( N54Fan ?) has backtested this crossover on daily charts. It might be a usefull help on some swing trades. Thanks for reminding me in what kind of thread I had posted...
Language, no, its a visual problem. I don't blame Lucrum for missing that, I too thought it was you who created that chart. Had to look again a few minutes ago to see that you were simply referencing someone else. Last year I did backtest those moving averages on stocks randomly selected from the nasdaq exchange. It failed miserably. I got a better result by going long after a green candle and short after a red candle and holding for 2.5 candles.
No problem, somehow I thought you asked about my reply to your posting of a chart with a moving average crossover. Maybe I responded to the wrong poster.
Using EMA crossovers in my experience has been most useful on really short time frames (15 min to 60 min) or really long time frames (monthly). This is assuming you are trading something inbetween (daily or weekly for example). The problem with EMA or SMA crossovers is that they are lagging indicators and will result in a lot of whipsaws during trendless environments. You said it worked quite well in the daily the last year but if you papertrade from 1-1-2011 to 12-31-2011 you will see you would have lost a lot of money. This would be especially bad from 8-2011 to early 12-2011 where huge market swings would have caused you to buy at or near tops and sell at or near bottoms. By the time you get a crossover you are well into the move. MA crossovers are by definition trend indicators and lag. The reason I say they are useful in longer time frames is because BIG long trends like the Bull market from 1995 to 2000 (in the chart you are referring to) are easily seen and it acts like a Locomotive train. Its hard to slow down and reverse course. Therefore, all pull backs should be bought. When you use it on a shorter time frame it acts more like an OSCILLATOR OF THAT LONGER TIMEFRAME. So in effect you are "hopping on" the Locomotive train when it slows down at the next station. The point being you only buy or sell in the direction of the dominant longer time frame trend. So look what would have happened if you ONLY went long every EMA crossover of the 60 min chart from 1995 -2000. You would only sell and go short when the monthly chart says to. Especially if you used margin you would have outperformed the market. I know there are many here that will say MA crossovers are a losing method but I will tell you they can be used fruitfully if you think of it in a commonsense approach. You MUST identify IF there is a trend FIRST before you use them. If there is no trend they are not as useful. Daily charts (in my mind...) do NOT show trends very easily. Weekly and monthly do. I hope that helps N54_Fan
Here is a simple example of what I am talking about. I posted earlier the chart of EMA 3 & EMA 9 crossover on MONTHLY charts as a simple and pretty reliable tool for identifying meaningfull longterm trends with relatively few whipsaws. Notice that from about 4-2003 to 12-2007 there was a HUGE UPTREND. (Notice that this would have also kept you out of the downtrend throughout ALL OF 2008 and up to 5-2009.) Below I have a 60 min chart from 4-2003 to 1-1-2004 (clear uptrend in monthly chart). Notice that the 60 min chart EMA Crossover now acts like a good oscillator to tell you when to BUY and remain long. This simple method assumes you do NOT go short in a countertrend move and only trade in the direction of the dominant trend. Edit: This is NOT the exact method I use to trade but has a big part of my method and system.