7 , Now the ea applying some Frontline .
ECN && STP && Scalping && Hedging - Broker Review - Oct 2009 24 replies
Alert on Fractals (MT) 2 replies
Zigzag & Fractals 10 replies
EA wanted for sound alert depending on fractals condition 2 replies
Fractals Edge Fx 3 replies
Disliked{quote} Mr_You, I really don't know what you are missing... Maybe the part about the "length" or "count" (total number of bars in zz leg) is not explained well enough and is missing from the docs. Lately it was brought to my attention. Sorry about that - I'm so deep in it that things seems obvious... Longer bars (pip wise) should be counted as 2-3 bars in the total zz leg bars count. The way I do it classify the bars by groups: Group1: bar size 1-5 pips = 1 bar Group2: bar size 6-10 pips = 2 bars Group3: bar size 11-15 pips = 3 bars etc... The numbers...Ignored
Disliked{quote} gg53, I think I've found a potential discrepancy. In the reply above you're referring to ZZ length in bars, but in your ZZ length prediction (http://www.forexfactory.com/showthread.php?p=7190111#post7190111) post you state the length is based on SET2 (the fractal set) which tells me the ZZ length is based on fractals. Are we checking ZZ lengths against both bars and fractals? If so is there a particular order in the logic for when these occur? Could you clarify or correct this please? Thank you very much!Ignored
DislikedCreating SET2: From the same historical data on M5 (2011-2012) we now create our ZZ and Fractals dataset. we apply 2 standard MT4 indicators, Fractals and ZigZag (13,8,5) on that historical data. An MT4 script that writes the results to CSV will be posted later on. For EACH ZZ leg we store its fractal sequence and the total length (bar count) of that ZZ leg. UP fractal is marked as "1". Down fractal is marked as "0". So a record of each ZZ leg should look something like this: {10011} - fractal sequence within that ZZ leg 12 - total length (bar count) of the ZZ leg. This is also stored in a database, and this dataset is called from now on as SET2. G.Ignored
Disliked{quote} Mine (and I also think his) Set 2 isn't a fractal set but a set which holds all information of a ZigZag leg. My set 2 looks like that e.g. Fractal Sequence 010101010 LengthValue(BarCount) 25 LegStartDatetime xxxxx LegEndDatetime yyyyy Direction 0Ignored
Disliked{quote} In dataset2 do we specify the direction of the zz leg with a "0" or "1"? As I don't recall gg stating that.Ignored
Disliked{quote} Lovely said . For me the most tricky part which hasn't been spoken off is from where to start the prediction and how to indentifiy a false prediction. If my last prediction was not correct I might be missing a few fractals of the next legs etc ...Ignored
Disliked{quote}The ZZ end-point is NOT my target. My target is the most probable trading SWING point. It's true that many times they are same but often it's not - because ZZ end-point is the HIGH or LOW of the bar and not its CLOSE. So looking for the "next bar in opposite direction" will give you different result than ZZ end-point - and it's not a "miss" or "error" as long as the trading entry point is better or equal or at least very near the ZZ end-point. G.Ignored
DislikedFractals, ZZ and the Renko bars (and don't forget the Pissing Dog...) The Renko version of the ForexGT_Arrow EA is finished (incl. Currency-Strength scanner and other supporting filters) . A major problem was the Backtesting of the Renko chart - which is always a problematic issue on almost all trading platforms. I think that we solved that issue too, and soon I'll post backtests results with at least 90% accuracy. Solving the Renko backtesting issue is a major cause of happiness here. Currently, the results of ForexGT_Arrow (Renko Ver.), EURUSD...Ignored
Disliked{quote} I apologize if this has already been answered. Just point me there. This seems to be my issue. Do you have any advice or are you able to describe how you go about looking for matches around the ZZ end-point? Or I guess another way to put it. If I have a ZZ match is there a number of bars previous and after that I should be looking for a SET3 match? Thanks!Ignored
Disliked{quote} Hi G, If you strip out money management (appears to be 10x leverage), how many pips did it make during these 2 months? About 480? Is this better than the average pip production on minute based charts?Ignored
Disliked@gg53 In an ideal world the direction for nextbar prediction is determined from last arrow drawn. But what about the first time you attach the indicator on your chart. No arrow to "check". Can you please tell how you determine the direction for this cause. Thanks in advanceIgnored
Disliked@gg53 In an ideal world the direction for nextbar prediction is determined from last arrow drawn. But what about the first time you attach the indicator on your chart. No arrow to "check". Can you please tell how you determine the direction for this cause. Thanks in advanceIgnored
DislikedOkay thanks Kilian19. next question for @ gg53 :-) Upper half of next screenshot shows the predicted arrows. Second half shows ZZ end points that are not repainted. {image} Question 1: The first 2 aqua arrows are "errors" because they are more than 3 bars after the MT4 ZZ? Question 2: If so we count the fractals and bars from last red arrow or green (latest confirmed ZZ high or latest confirmed ZZ low)? Thanks in advanceIgnored