I don’t see where version 1.3 has any parameter for setting the zzi period or is that in version 1.4?
Ikari - when you first start with an EA it has default settings. Paremeters for the bot are established by the best settings have not yet been determined. With that you put the bot in backtesting mode with Mt4 platform tester or equivalent. After running backtests you will get a buch of possible optimizations. After you investigate each of those setups individually you can save those settings as a “.set” file. Then in the future when you put the EA on a chart (live or demo) you can load those set file settings into the EA and it will trade by those settings. This EA just went through a major upgrade so all the set files we had in the past are probably not very good with the new features available to us. So everyone needs to pick a currency and time frame and run short term optimizations. Pick the best settings and run longer term backtests with those to validate the optimization…then of course it has to go on demo then on live account to really be worth its weight in gold.
I have traded system today with your .set loaded and check this strange trade… Very good SELL entry that ended with loss after missing all profit…
Can You comment this?
Let me say this right now. The EA is an exceptional trade picker but a lot weaker at trade closing IMO. In live trading I have been closing trades when profits seem to peak or direction seems to be hinting of changing against my position. I watch the ZZ and fractals pretty closely. I don’t like to hold trades through retraces or weekends, that’s just not me. The EA will hold through numerous retraces of a position depending on the trendline, fractals and stop loss.
In looking at your chart you posted, with out fractal indicators it was hard to see what happened. Put fractals on there and I think you will see. After entering the trade it shot down to a low fractal. Then the market moved against you upward going to a lower high fractal than the previous one. IMO the bot says you are below your previous fractal SL therefore still safe and it will move a new stop loss down to that new fractal. HOWEVER, the market never moved down far enough to get away from that fractal and conintinued to move up which would prevent the bot from establishing a new stop loss there unless you were asking it to be at least 50 pips above that point. There is probably an error in your error log stating SL move failed. At that point it looks like it may have either hit your old fractal SL if it was still there and not cancelled, or your BE point depending on what you set it at an if you have that feature, or your ultimate shirt hard SL. This is why I still like to manually watch the trades as there are quirky situations that arise that will stump any bot out there. You just can’t code every possible outcome that may come at you. Just like I don’t leave my kids unsupervised, I espesciallly don’t leave someone trading my money unsuprevised either.
Great feedback guys, the new version of the EA will have a drop down field that will allow you to choose pre determined .set files to make thing really easy, so for Example you might choose:
EUR/USD 15 Minute
EUR/USD 30 Minute
etc.
etc.
this will pull .set info from the server to give you the latest and most optimized set for that pair and time frame. We should have this version ready within 3-5 days. This will avoid a lot of headache. And as users help us optimize, we can upload their .set files onto the server for all of us to use.
Thank you for information. The problem is that i dont have time to manually control trades executed by EA, i am searching for fully automated robot. Is there a way to tune SL for this .set?
Ikariux17, that’s what we are trying to do here. We all dream with the perfect full automatic and eficient robot adding pips.
Here we have the rare oportunity to test this EA (that already works) for free:D
Just as Rome was not built in one day, neither can a bot be judged by one trade.
As others have mentioned, there is no such thing as a holy grail or no loss perfect bot. All bots have some extended forward testing statistics that say it should win X% of the time and Y pips. We have to live in those statistics and understand that there will be losing streaks but if the statistics are valid we will prevail in the long run. Too many traders runs some tests and see that overall a bot does real well then begin testing it only to abandon it after it has a few losing trades to go in search of a better grail. It is our trading nature to try and do better and to perfect before we really understand what we have. I’m not saying to never abandon a bot, but I’m saying you really need to test it and understand it before you will really know if it is the right one for you.
Again as I stated before, I wouldn’t hand someone I don’t know well all of my money and say go trade it and have fun. You should really know your bot before trading away, but even then I would still keep an eye on it.
Hello, A new version of the Premji_ea will be available this week, I have setup a very profitable setting using the 1 minute EUR/USD chart. Basically the new EA will pull these setting from a server where they have already been optimized for the user, simply drag the EA on the chart and choose the correct pair and time frame and the rest is already optimized for the user.
Basically for each new month, we have optimized the setting for the EURUSD 1 minute using historical data going a specific number of months back. So a new set will be posted Jan 1st and has been optimized using a special algorithm our team have been working hard to establish. We have found going back 12 months, for example Jan 2010, we created a set. using data from a predetermined period of time in a specific manner, not including Jan 2010 of-course. so we created a .set for Jan 2010 using data from a few months prior to Jan and found that you can get at-least 250-300 pips a month from EURUSD 1 minute, we did this for all of 2010. so 1 for Feb. 2010, again using this special optimization .set used to determine a perfect set for Feb. 2010 using data from a few months prior to Feb. and again we generated the same consistent 250-300 pips for Feb. using a .set created using data prior to Feb. We did this for all of the months in 2010 and it worked like a charm. Basically here is what the user can expect.
At the beginning of Each new month, so the next will be Jan, 2011
my team will be using data going back a specific number of months to create the most profitable .set file with minimal draw-down for the EURUSD 1 minute. The values of this .set file will be uploaded to our server, on Jan 1, when you drag the new Premji_ea onto the EURUSD 1 minute chart, the .set values will automatically be pulled from our server so the user will not need to enter or optimize these values.
So as I said we have done this for each month is 2010 just to make sure our algorithm has worked, and it has very consistently. Every month on average this system produces 250-300 pips for EURUSD 1 minute alone with approx 1.5-1.75 profit factor and 8-10% DD. So if you are trading 1 lot with a 10k account, you can expect $2500+, the EA also takes about 50-60 trades a month.
So we look forward to getting this up a going within a week or so. After this the plan will be repeating this process for other time frames so we can eventually get to 1000 pips or so/monthly with 3-4 EA’s running on 4 charts. At the beginning of each month, the user will place the EA’s on their charts and the optimized .set values will be pulled from the server making things very simple.
An issue we may have is that these .set files were created using FXCM’s historical data and may differ from your broker but i guess a demo on other brokers account will reveal this by the end of Jan.
Here is a EURGBP M5 set that looks really good for the last three months. People may want to run a few more backtests on it before putting it on anything real.
Look forward to the next updated version. We are making progress.
Here are two potential set files for USDJPY on M5.
A file has more trades but less profit.
B Takes far less trades but has not had any losers.
Again, due to the limitations of my testing computer these were only tested in small segments in back tests. One may want to run some extended trials with them before moving forward.
Hi Guys, here is the new EA, simply place it on the EUR/USD 5 minute chart, change the USmarketclose to 23:59 and the rest is pulled from the server. I will attache the graph and report.
Basically 10K turns into close to $90,000 in 12 months. We have back-tested this 5 minute EA for 12 months and it has done really well.
Can someone point me in the right direction? After attaching EA to chart I get error messages for not being able to open Mega_Trend and ZeeZee.
In addition do I need to set GMT time (my broker is GMT+2) and then just change US Market Close to 23:59? Did both and tried to run backtest with no success.
