Lets Automate a Simple Renko Strategy! | Page 45 | Forex Factory

Options

Go to
Permalinks
Print Thread

Related Instruments

EUR/USD

Similar Threads

I will code & automate your strategy for free 110 replies

Should I Automate My Strategy? 10 replies

WRB/MC strategy:Coder needed to semi-automate my strategy 11 replies

My Simple System (Can we automate this system?) 96 replies

How can I fully automate my strategy? 2 replies

Attachments: Lets Automate a Simple Renko Strategy!
Exit Attachments

Lets Automate a Simple Renko Strategy!

  • #881
  • Edited 6:05am Apr 8, 2015 5:42am | Edited 6:05am
hi, please find the results of the latest EA core backtest attached ( picture 1 ). voila, it's a backtest, a lot of trades, no filtering, just improved exits.


based on these entries / exits i will apply the filter shown in picture 2. entry is close price of the bricks marked with an arrow. the EA does not need more than 1 consecutive brick after entry to close without losing money. this logic is already implemented in the backtest result. picture 2 shows the market as of now, no backtest, live. i hope i can implement by tomorrow so that i can see first forward testing results on friday.

have a good day, v.
Attached Image(s) (click to enlarge)
Click to Enlarge Name: ART EA Core Results.jpg Size: 150 KB
Click to Enlarge Name: BBB.jpg Size: 250 KB
  • #882
  • Apr 8, 2015 5:50am Apr 8, 2015 5:50am
niceee ! well done, thank you guys math trader and verified for yuor great work
  • #883
  • Apr 8, 2015 9:15am Apr 8, 2015 9:15am
Quoting verified
hi, please find the results of the latest EA core backtest attached ( picture 1 ). voila, it's a backtest,

Refresh us again, how is your current EA working out based on the Backtests?

Any sort of win:loss ratio and an equity curve would be AWESOME.

Which, again, I am pretty sure you cannot provide, because based on all of your other backtests, the forward testing sucks the big one...right?
Money Can't Buy Happiness. Poverty Can't Buy SHIT! You Choose!
  • #884
  • Apr 8, 2015 10:10am Apr 8, 2015 10:10am
Magix, have you ever thought about a constructive posting in the last period of time ?

Not saying, that something does ( could ) not work but how it works ? Anyway, I will explain to you how Renko Test Data must be seen in the context of backtesting to bring you on the same page.

When a programmer converts time bars to renko bars, he/she writes one record for every renko brick into a file. This Record contains a time stamp, the H, O, C, L Prices and a tick count for the older MT4 history file structure. Have you ever wondered why the price jitters when you visual backtest ? This are random fake moves from the strategy tester. Now, what does this mean ?

Saying something is not reliable is not good enough, we don't want to know the problem, we want to know the solution. And the solution is simple. Never compute something in your strategy based on the open Candle[0], the data shown inside this candle ( the strategy tester knows already how it looks like but wants to give you some live experience before showing it ) is unreliable, the data of this Candle once it's closed is more than reliable enough for what we are doing here. So it depends on your code whether strategy testing provides random or reliable results.

I hope this helps you for better understanding and as always, If I can help, just ask.

Have a nice day, V
  • #885
  • Apr 8, 2015 11:15am Apr 8, 2015 11:15am
Quoting verified
Magix, have you ever thought about a constructive posting in the last period of time ? Not saying, that something does ( could ) not work but how it works ? Anyway, I will explain to you how Renko Test Data must be seen in the context of backtesting to bring you on the same page. When a programmer converts time bars to renko bars, he/she writes one record for every renko brick into a file. This Record contains a time stamp, the H, O, C, L Prices and a tick count for the older MT4 history file structure. Have you ever wondered why the price jitters...
So, wuh....

Ya still got nothing, hey?

If you have a working model of your EA that can run a backtest, it can actually run live as well...if the performance doesn't match the backtest, what does that tell you?

Your EA from start of week failed. You based success on the Bactest Reports.

Anything you have backtests on, should be able to produce a live equity curve based on "Actual" market.

Yet you fail to provide anything but pretty pictures of worthless data.

Please explain how this is constructive?

As far as I can see, you are still stuck in stupid mode, and failing to see the forest through all those pesky trees.
Money Can't Buy Happiness. Poverty Can't Buy SHIT! You Choose!
  • #886
  • Apr 8, 2015 11:20am Apr 8, 2015 11:20am
Hi Math,

I have been testing version 2.9 today and it looks sometime trailing SL doesn't work for me well. I set TSL as "Every3Candles". As you can see on the picture (GOLD) so this situation happened but SL is still on the start position, last swing.

Thanks for help
Attached Image
  • #887
  • Apr 8, 2015 11:25am Apr 8, 2015 11:25am
Quoting verified
hi, please find the results of the latest EA core backtest attached ( picture 1 ). voila, it's a backtest, a lot of trades, no filtering, just improved exits. based on these entries / exits i will apply the filter shown in picture 2. entry is close price of the bricks marked with an arrow. the EA does not need more than 1 consecutive brick after entry to close without losing money. this logic is already implemented in the backtest result. picture 2 shows the market as of now, no backtest, live. i hope i can implement by tomorrow so that i can see...
Nice work. I like systems where isn't so big numbers of profit trades like 90% etc. More credible is for me this ratio, 55% in profit, 45% loss but if you can stay in profit for longer time.;-) I hope you will be successful and share EA for testing.;-)
  • #888
  • Apr 8, 2015 1:48pm Apr 8, 2015 1:48pm
Hello,

Can any body tell me what is the final version of EA on this Strategy please ?
  • #889
  • Apr 8, 2015 1:50pm Apr 8, 2015 1:50pm
Quoting dafa
Hi Math, I have been testing version 2.9 today and it looks sometime trailing SL doesn't work for me well. I set TSL as "Every3Candles". As you can see on the picture (GOLD) so this situation happened but SL is still on the start position, last swing. Thanks for help {image}
I have tested the trailing stop functionality on standard Renko charts. It may not work as expected on Median Renko charts.
Trading is the hardest way to make easy money...
  • #890
  • Apr 8, 2015 2:05pm Apr 8, 2015 2:05pm
HI Math Trader, i am tetsing every day your EA with real time charts to see how it works, and it is really great, the only thing is when price is extended and there is a small bounce from price and ea continue selling in an oversold or aver bought, what i see is that the ea gives great entries when all the trend start but at the end is where i see lossiing entries. i am trying to dleete those entries with stocastich and following i see that when the trend is down and the ea start selling in 80 sto territory is great but i stop it when the sto is under or close to 20 to avoid not strong part of the trend.

i dont know what do you think if you see that the Ea starts to lose there, is just a comment and all your comments i really apreciate

best
  • #891
  • Apr 8, 2015 3:04pm Apr 8, 2015 3:04pm
Quoting aledipi
HI Math Trader, i am tetsing every day your EA with real time charts to see how it works, and it is really great, the only thing is when price is extended and there is a small bounce from price and ea continue selling in an oversold or aver bought, what i see is that the ea gives great entries when all the trend start but at the end is where i see lossiing entries. i am trying to dleete those entries with stocastich and following i see that when the trend is down and the ea start selling in 80 sto territory is great but i stop it when the sto is...
I'm also adding/deleting indicators to the Renko chart the EA attached to, trying to find some method to reduce the number of losses. Stochastic oscillator is one of my early choices, the problem is to select the input settings that are best for all the market conditions. But are there such magic input settings? if so, what are the values?
Trading is the hardest way to make easy money...
  • #892
  • Apr 8, 2015 3:33pm Apr 8, 2015 3:33pm
Quoting MathTrader7
{quote} I'm also adding/deleting indicators to the Renko chart the EA attached to, trying to find some method to reduce the number of losses. Stochastic oscillator is one of my early choices, the problem is to select the input settings that are best for all the market conditions. But are there such magic input settings? if so, what are the values?
Although, not a Renko Trader, I am using a 32, 5, 3 from the H1 on one of my other traders that works with a profit trail.

Synthesized 8, 5, 3 for an H4.

What this does is allows the trades to build and grow, but restrictive enough to keep you from buying tops and selling bottoms.

Filtering through an additional time frame has actually worked out well for a couple of my current traders.

Filtered by the day, although a basket trader...probably not what you are looking for:
http://widgets.myfxbook.com/widgets/1208479/large.jpg

This is the regular equity millipede, filtered on a synthetic H4 stochs.
http://widgets.myfxbook.com/widgets/1188215/large.jpg



The MTF stochs are a little difficult to apply to a renko chart because of the refreshing issues, but if you compare your brick size to a different time frame, might come up with the answer you are looking for.

One of the new traders I was working on this week was to be a true H4 8, 5, 3, equity millipede, but the issue was the number of pairs and load from other terminals. Will have to get a whole other vps/puter up and running for this one.

Just a couple of ideas.

Money Can't Buy Happiness. Poverty Can't Buy SHIT! You Choose!
  • #893
  • Apr 8, 2015 4:21pm Apr 8, 2015 4:21pm
Quoting Magix
{quote} Although, not a Renko Trader, I am using a 32, 5, 3 from the H1 on one of my other traders that works with a profit trail. Synthesized 8, 5, 3 for an H4. What this does is allows the trades to build and grow, but restrictive enough to keep you from buying tops and selling bottoms. Filtering through an additional time frame has actually worked out well for a couple of my current traders. Filtered by the day, although a basket trader...probably not what you are looking for: {image} This is the regular equity millipede, filtered on a synthetic...
Thanks Magix! I'll go for (35,5,3) settings and test them. However, frankly speaking, I do not believe that there are such magic settings for any indicator ;-)
Trading is the hardest way to make easy money...
  • #894
  • Apr 8, 2015 4:37pm Apr 8, 2015 4:37pm
Quoting MathTrader7
{quote} Thanks Magix! I'll go for (35,5,3) settings and test them. However, frankly speaking, I do not believe that there are such magic settings for any indicator ;-)
I am using the 20 on my coin toss as a reversal point...

You can check the stats on that as per the trade pair on a 10 pip brick.

IF you are not buying when it is in overbought, or selling when it is in oversold, you may not make the full range of the trade, but you will most likely not be buying tops and selling bottoms.
Money Can't Buy Happiness. Poverty Can't Buy SHIT! You Choose!
  • #895
  • Apr 8, 2015 4:44pm Apr 8, 2015 4:44pm
Quoting Magix
{quote} IF you are not buying when it is in overbought, or selling when it is in oversold, you may not make the full range of the trade, but you will most likely not be buying tops and selling bottoms.
I totally agree with this point!
Trading is the hardest way to make easy money...
  • #896
  • Apr 8, 2015 8:03pm Apr 8, 2015 8:03pm
Renko - Ichimoku Cloud combo in action. No enhanced exits, no trailing stops, not optimized. Shows good results already, I'm going to optimize now. Nothing is calculated with the forming candle, testing is based on closed candles without exceptions.

FYI and have a nice day
Attached Image (click to enlarge)
Click to Enlarge Name: NNN.jpg Size: 293 KB
  • #897
  • Apr 8, 2015 8:15pm Apr 8, 2015 8:15pm
Quoting verified
Renko - Ichimoku Cloud combo in action. No enhanced exits, no trailing stops, not optimized. Shows good results already, I'm going to optimize now. Nothing is calculated with the forming candle, testing is based on closed candles without exceptions. FYI and have a nice day {image}
Money Can't Buy Happiness. Poverty Can't Buy SHIT! You Choose!
  • #898
  • Apr 8, 2015 9:04pm Apr 8, 2015 9:04pm
Quoting verified
Renko - Ichimoku Cloud combo in action. No enhanced exits, no trailing stops, not optimized. Shows good results already, I'm going to optimize now. Nothing is calculated with the forming candle, testing is based on closed candles without exceptions. FYI and have a nice day {image}
hi verified

observed from the picture, you use 1 candle rising instead of 2 for entry, exit is the same. is that right? and Ichmomoku for trend confirm?

thanks
patience is a virtue
  • #899
  • Edited 9:36pm Apr 8, 2015 9:14pm | Edited 9:36pm
hi blueseahorse,

this is not a classic Renko entry ( 2 cons ... ). i enter when the price - the prev candle - closes completely outside of the Ichi cloud and is not touching the cloud. long above the cloud, short below the cloud. brick size 8 pips, initial SL 15 pips. nothing I have invented, i just make the EA . currently working on better exits ( again ). the expensive exit is the killer of the OP's strategy, i think.

will need a time filter for this strategy as well.

hv a nice day !
  • #900
  • Apr 8, 2015 9:46pm Apr 8, 2015 9:46pm
Quoting verified
hi blueseahorse, this is not a classic Renko entry ( 2 cons ... ). i enter when the price - the prev candle - closes completely outside of the Ichi cloud and is not touching the cloud. long above the cloud, short below the cloud. brick size 8 pips, initial SL 15 pips. nothing I have invented, i just make the EA . currently working on better exits ( again ). the expensive exit is the killer of the OP's strategy, i think. will need a time filter for this strategy as well. hv a nice day !
i see, good luck with the EA, looking forward to the results
patience is a virtue
1 trader viewing now