Parabolic SAR robot Review

[Batchboy]

Originally Posted by PipCop

I prefer the 3 months range for demos, because by then I can usually tell if its going to perform or not. Once it goes live, though, I have no time limit - per the Review FAQ - robots run until they fail.

cool.

Well, I'm running a live micro account anyway if anyone wants to "listen-in" on that progress for however long. So far for its few days existence it's running similarly to demo.
jb

[PipCop]

Well, how about this: I have created a new area in the site for Member Journals:

http://www.PipCop.com/forums/member-trading-journals/

You are welcome to start a new thread there, and post updated statements as you please.

[Batchboy]

Uh-oh! Losing more now than all gain last month. You sure David this isn't "optimal" for certain market conditions? My micro-live has gone nearly a few bucks below the initial $18 too!
This is what I feared of a simple one TA kind of EA.
jb

[david.jukl]

Hello Batchboy! SAR does´t work well in side trend . It works perfect in uptrend, downtrend, but not when there´s no trend .... Any suggestion for recognition of trend or modification? That´s why I test this simple stategy here

[Batchboy]

Originally Posted by david.jukl

Hello Batchboy! SAR does´t work well in side trend . It works perfect in uptrend, downtrend, but not when there´s no trend .... Any suggestion for recognition of trend or modification? That´s why I test this simple stategy here

David,
Hmm, I guess since recently EA's are now required to have gone thru live testing, but you got in under the radar being that testing for yours is now over 3 months, ok, no prob.
If I had an Indicator that could indicate trend , I'd be a millionaire (new movie name, instead of "slumdog millionaire" it be "FX millionaire").
But maybe something that comes close could help so that your P.SAR kicks-in at whenever time your, there are some free indicators at mqlsoft.com, look at charts with one of those indicators and your P.SAR indicator and see if there is a method that can be programmed.
That's what I did with upcoming new EA being programmed by 1ka, ("Invisible FX Landscape", new thread regarding that). I started anew by facing chart and putting up some indicators and conceiving what could be needed. One thing might be possible for you is retracements as a kind of "trend", you know where they always are, after a trend , just look at chart with RSI &/or RVI and see that during the trend these are "high" in one direction or the other so you would have a set-up = 1 (what I'm having done in new EA), thing is it's not always some jolly good reverse trend , but it could be better than 80% of the time "in a pig trough" wriggling around in a consolidation period over and over! Follow? I say "could be better", take a look at chart, and see if you want to "take your chances" in that route, because retracements may only be a a small wriggle half the time, but it's better than 80% of the time (heard that is ballpark % of time in consolidations vs. trend ?) sloshed around in consolidation periods, no? A thought.
Amazing what your P.SAR had done this well being only 1 indicator and needing be applied only in 20% of the time periods re: "trends", pretty amazing considering! But despite that wonderment, it's best if a method can be arrived at.
jb

[velocity]

Hi David,

Great EA and if it can be optimized to trade during the strong trend , that would be awesome! Of course, as Batchboy says, we would be instant millionaires if we can do that. Until then, I guess we will have to manually monitor for when to turn it on and off.

By the way, do you have any live statements that we can currently look at?

TM

[Batchboy]

Originally Posted by velocity

Hi David,

Great EA and if it can be optimized to trade during the strong trend , that would be awesome! Of course, as Batchboy says, we would be instant millionaires if we can do that. Until then, I guess we will have to manually monitor for when to turn it on and off.

By the way, do you have any live statements that we can currently look at?

TM

hi Velocity,
I have a live that's been running for awhile, link at lower post:
http://www.PipCop.com/forums/member-trading-journals/129-live-parabolic_sar-18-01-initial-equity.html
jb
PS: I strongly recommend doing this P.SAR as I mentioned, on a retracement via strength indicators, because those are LIKE knowing when SOME KIND of trend is about to happen......right after the strength indicated actual trend move. I have upcoming Invisible_FX_Landscape doing that just fine (well, at least back-tests show so far as good), two indicators to look for crossing of the beginning of a move, and RSI/RVI as alternative choice set-ups as to when to partake of said crossings. Not sure if David can also partake of 1ka's helpful programming work (but it be behoven idea), got his help because I made a thread from get-go that attracted his attention. It often is said back-tests aren't any good for showing future results, but I think I have found it mostly has to do with not taking great care in seeing back-tests/optimizations are replicated via live, and this I do as a careful study any more. Given that then the average optimized performance for previous 3 months SHOULD (!?) be projected into average performance walking forward, no? Other opinions? Can't say I'm any sort of expert YET(!!)_, but hell, if I keep this up I may be (but only as a group member with Pros like 1ka et al)
jb

[PipCop]

Please note : The trade placed at 2009-02-17 20:35 BUY was an error. I moved this robot to a new VM and forgot to set the mode to SHORT only. I will leave it open to see if it can eek out a profit.

[ICUR2Ys]

Hi all,
I am very new so this may be a stupid suggestion but here goes.
Why not add ADX. ADX = 20 or less not trend . ADX = 30 or more = strong trend .

[david.jukl]

At the end of this month I would like to mention that also this month was profitable. Final monthly loss of 68USD was caused by bad trade 2009.02.17 20:35buy. Strategy is designed to do short trades only....