request: RSI + ATR | Page 2 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Correct number of weeks for ATR and hour ATR 1 reply

Request: ATR Indicator Made Up Of Data, Not Visual 11 replies

RSI cross indicator when rsi=50 19 replies

RSI (Fast) above/below RSI (Slow) displayed as a Histogram 5 replies

RSI, MA, BOLL or RSI, BOLL MA diferents results 2 replies

Attachments: request: RSI + ATR
Exit Attachments

request: RSI + ATR

  • #21
  • Aug 16, 2006 10:32am Aug 16, 2006 10:32am
I know. But I am wondering if the EA "forgets" the trade if the PC is rebooted or the connection break temporarily.

Quoting wackena
The way I think it works is when an EA places order/s with s/l & t/p printed, the server will process the order/s to conclusion at either s/l or t/p without the need of being connected to your PC.

On order/s that are placed without a s/l or t/p printed, your PC needs to be connected to allow the EA to process the order/s to conclusion. If PC is not connected, the order/s can run rampant in either direction.

Wackena
  • #22
  • Aug 16, 2006 12:25pm Aug 16, 2006 12:25pm
Quoting Yardie
I know. But I am wondering if the EA "forgets" the trade if the PC is rebooted or the connection break temporarily.
Yardie, to test, I just shut down the demo running this EA and restarted. It seems to be OK, no data lost.

Wackena.
  • #23
  • Aug 16, 2006 12:35pm Aug 16, 2006 12:35pm
Wackena what pairs do you have it on? I have a profitable trade in eurusd at 53 pips. But I am not sure if the EA will close it automatically or not. The last thing I need is a profitable trade to turn into a losing trade.



Quoting wackena
Yardie, to test, I just shut down the demo running this EA and restarted. It seems to be OK, no data lost.

Wackena.
  • #24
  • Aug 16, 2006 4:15pm Aug 16, 2006 4:15pm
Quoting Yardie
Wackena what pairs do you have it on? I have a profitable trade in eurusd at 53 pips. But I am not sure if the EA will close it automatically or not. The last thing I need is a profitable trade to turn into a losing trade.
Yardie, there is code in the EA to close orders. So far in backtesting, it has worked OK. Pairs trading are eur/usd, gbp/usd. usd/jpy, usd/chf, usd/cad, aus/usd, eur/jpy, eur/chf, eur/gbp.

On another subject, I've noticed that on FXDD & North Fiance Demo, the Money Management routine (UseMM) will not trade when lot size is below 0.1. As example: Deposit $5,000, Risk 2%. This will calculate lot size to 0.1. If loses happen right away and lot size is calculated to below 0.1 the trading stops.

My IBFX demo worked OK with lot sizes under 0.1.

Wackena
  • #25
  • Aug 16, 2006 6:10pm Aug 16, 2006 6:10pm
i tried this on the 1Hr backtesting eur/usd and did not get 1 trade logged.

is there a trick to this backtesting?

the graph and summarys are blank
  • #26
  • Aug 16, 2006 6:18pm Aug 16, 2006 6:18pm
Quoting HAGeek
i tried this on the 1Hr backtesting eur/usd and did not get 1 trade logged.

is there a trick to this backtesting?

the graph and summarys are blank
Try UseMM=0. Lots=0.1 or above. See if it trades now.

Wackena
  • #27
  • Aug 16, 2006 7:00pm Aug 16, 2006 7:00pm
Here is the original RSI strategy. This is only part of the full strategy.

http://www.traders.com/Documentation...OREXfocus.html

TimeFreedom
  • #28
  • Aug 20, 2006 4:28pm Aug 20, 2006 4:28pm
hmm... dunno what i am doing wrong.

i set a date to run from and to. select the EA. adjust input variables if needed.. a time frame.

then hit start.

the bar graph below doesn't move to right and no results in the results or graph tab.

so whats up?
  • #29
  • Sep 5, 2006 12:16pm Sep 5, 2006 12:16pm
Hi guys,
I have read this thread and am interested in this strategy as well. I have not however been able to find the whole article anywhere in it's entirety. I have seen some parts of it. Would someone please post the text of the article? I am a pretty experienced programmer and would like to see if I can do a good implementation of the strategy.
  • #30
  • Sep 5, 2006 12:19pm Sep 5, 2006 12:19pm
I should have siad that I am wondering if there is any more to the article than what has been posted? In the first post Burger King you mentioned percentages of the ATR, did you calculate those on your own, or did you see those as part of the article you read?
  • #31
  • Sep 5, 2006 12:47pm Sep 5, 2006 12:47pm
Sorry, one last thing,
I need to get my brain in gear before I hit the post button. I just wanted to mention also that I think you guys have done a lot of good work here on this, it's the main reason I have become so interestred to try it out. Good job Wackena on the EA.

I have not been able to get the same results as V2 with my dealer (interbankFX) but I'll look a little closer and try to find out why.
  • #32
  • Sep 5, 2006 5:42pm Sep 5, 2006 5:42pm
Quoting jhfx1
I should have siad that I am wondering if there is any more to the article than what has been posted? In the first post Burger King you mentioned percentages of the ATR, did you calculate those on your own, or did you see those as part of the article you read?
Its from the article. The complete article could be found on a TASC august issue (paperback - not online) magazine. The web version is an excerpt unfortunately.
  • #33
  • Jun 22, 2007 11:55am Jun 22, 2007 11:55am
I'm glad to see that this thread exists, I was going to start making this EA for MT4. Maybe I can join in on this EA's development.

Here are some of my questions and contributions:

Let's say theoretically that I am trading the EUR/USD on the 1 hour chart and the RSI crosses 50.

I don't understand Buying at 15% of daily ATR above previous high

If I'm on the hourly chart where do I get the ATR? From my hourly chart (current bar or previous bar), or from my daily chart (current day or previous day)?

Do I add the 15% of ATR to the current market rate, high of previous bar on my hourly chart, high of previous bar on my daily chart?

I attached a little something that should help us as well. Sorry about its crappy quality but it was originally around 3MB and I had to compress it. Admin let me know if there is a prob with attaching articles and I will remove it.
Attached File(s)
File Type: pdf article_new_small.pdf   485 KB | 1,122 downloads
  • #34
  • Jul 17, 2007 10:52am Jul 17, 2007 10:52am
Any update on this EA?

Thanks
Patience, let the market come to you!
  • #35
  • Nov 9, 2007 3:20am Nov 9, 2007 3:20am
Quoting BurgerKing
According to the article, that RSI + ATR discipline was backtested for 3.5 years using a startup capital of $1000 and ended up with more than $60,000 of 60x. And this was using 1 lot sizes from day 1 to 3.5 years later. The return can be higher if the script scales up based on the equity at hand.

Backtest result I posted over at VT forum:


PS. its RSI(21) and ATR(21)
Does anyone know of a way to download historical data into VT. I seem to get about one month of data on an hour chart. How can I get more to back test properly??
Katte, kabuto no o wo shime yo. Size doesn't matter, leverage is the killer
  • #36
  • Mar 26, 2011 1:24am Mar 26, 2011 1:24am
Quoting neta1o
I'm glad to see that this thread exists, I was going to start making this EA for MT4....
The answers are within the article that YOU posted!

*** Article says "2. Buy at 15% of daily ATR above previous high (hourly, 15-minute, and so forth)"

To me, the ATR always means a number from the DAILY chart, only the number of days change, that is normally 14 but for this system, he has chosen to use 21 for the reasons explained in the article.

Now, hit me with it !!!
Canadian by birth Australian by choice
  • #37
  • Mar 26, 2011 1:29am Mar 26, 2011 1:29am
Quoting TheSnowman
The answers are within the article that YOU posted!

*** Article says "2. Buy at 15% of daily ATR above previous high (hourly, 15-minute, and so forth)"

To me, the ATR always means a number from the DAILY chart, only the number of days change, that is normally 14 but for this system, he has chosen to use 21 for the reasons explained in the article.

Now, hit me with it !!!
I don't think he'll hit you. He hasn't visited the forum since Aug 2007.

Like you, I've only ever thought of ATR as a daily true range.

Anyway, it would be interesting to see if the original strategy tests just as profitably 4 years later.
  • #38
  • Mar 26, 2011 3:21am Mar 26, 2011 3:21am
I'm sure it still has the same potential; did you have a look at the article? The 4 H equity curve sure looked GREAT !!

I might watch just the EUR-USD on the 4 H chart next week, won't take much time to manage that, and there is a set up in real time right now, no entry yet.

Here is a chart and my price calculator, just rounded for convenience
Attached Image(s) (click to enlarge)
Click to Enlarge Name: 3_28_2011_eur.1.jpg Size: 122 KB
Click to Enlarge Name: RSI_CALCULATOR.png Size: 48 KB
Canadian by birth Australian by choice
  • #39
  • Mar 26, 2011 7:08am Mar 26, 2011 7:08am
I read the article. Entry rule #2 (for a buy) reads "Buy at 15% of daily ATR above previous high (hourly, 15 minute, and so forth)".

Suppose I'm running the method on an H1 chart. Does this mean (1) "Buy at 15% of H1 ATR above previous H1 high" or (2) "Buy at 15% of D1 ATR above previous H1 high"?

To be consistent with Rule #3, it seems that (1) would be correct. But if we read on, the comment "The strategy tested profitably on intraday charts...... The values would likely need to be altered for trading on a daily chart" suggests to me that (2) is more likely to be the correct interpretation. I see the writer of post #12 also noted this ambiguity. Getting this wrong may or may not account for the poor backtest profitability noted in posts #8 and #9.

I'm curious to know why this simple, indicator-based approach would significantly outperform any similar trend-following system. As soon as the writer talks about optimizing values, I become nervous, as I'm acutely aware of the dangers of curve fitting indicators. Also, I see that the equity curves are generated from around 500, and 220 trade samples, respectively, and the stats in figure 6 are the result of a mere 213 trades. The PF of 2.26 seems to be bordering close to the "too good to be true" category. I'd like to see a test run over an absolute minimum of 2,000 trades and 5 years' worth of changing markets. My experience is that very few automated systems will deliver a PF > 1.2-1.3, if tested over a suitably large sample.
  • #40
  • Mar 26, 2011 8:43am Mar 26, 2011 8:43am
Quoting hanover
I read the article. Entry rule #2 (for a buy) reads "Buy at 15% of daily ATR above previous high (hourly, 15 minute, and so forth)".

Suppose I'm running the method on an H1 chart. Does this mean (1) "Buy at 15% of H1 ATR above previous H1 high" or (2) "Buy at 15% of D1 ATR above previous H1 high"?

To be consistent with Rule #3, it seems that (1) would be correct. But if we read on, the comment "[i]The strategy tested profitably on intraday charts...... The values would likely need to be altered for trading...
The values would likely need to be altered for trading on a daily chart

Well, actually, that makes it clear for me, he is saying that this was tested on 1 H & 4 H charts, so the ATR used was DAILY as there was never any intention to use the system on daily charts.

In regards to values, most likely he means -
1. The RSI period; 21
2. The % used for entry and stops

In fact, he says this " As mentioned, you can trade different time frames. If you trade on a daily chart, then you will need to widen your stop, perhaps to 100% of ATR."

I do not have the means to back test, so will have to take my chances.

Still, interesting to have a look at the daily RSI, see chart, gave an excellent divergence back in January, along with a 50 crossing signal, 3 days where you could enter that signal, #1, then watch all pips come right back to starting point before taking off.

Too bad, the chart is not telling me what price is doing next, not yet anyway!
Attached Image (click to enlarge)
Click to Enlarge Name: eur_daily.png Size: 18 KB
Canadian by birth Australian by choice
0 traders viewing now