AlligatorEx. - page 9

New comment
 
ZZZEROXXX:
Please send the indicator here, I will try it too. If it's different from this one https://www.mql5.com/ru/code/9828

The bicycle is ready. Of course there is no alligator there, I wanted to see what I could squeeze out of one. Trade on channel breakdown.

In addition to general settings, there are signal settings in the settings:

SignalByCandleClose - wait for the close candle above/below the channel

BreakevenPtsFilter - we can set the value of breakdown in points, if 0, it will be ignored.

ChannelSlopeFilter - use or not the channel slope filter for making a deal

ChannelSlopeFilter - to use or not the channel slope filter for making transaction

SlopeFilter_DirectUndirect - if the previous parameter is enabled, then (by example, selling) true - selling at breakdown of the sloping channel, false - selling at breakdown of the channel sloping upwards

Results of runs with the optimizer:

EUR/USD 1H Alpari Benchmarks
01/01/2010-01/01/2011

1) Breaking the previous candle's channel boundaries Profit 610, DD 13.78%, TT 192, Math 3.18
2) The previous candle closed outside the channel limits, calculated on this candle Profit 707, DD 11.86%, TT 162, Math 4.37
3) Closing the previous candle outside the channel boundary calculated on the previous candle Profit 1124, DD 12.07%, TT 152, Math 7.40
4) Variant 1 + filter for breakdown of the channel 40pts (optimization 5 - 50) Profit 1129, DD 10.3%, TT 50, Math 22.58
5) variant 2 + channel breakdown filter 25pts (5 - 50 optimization) Profit 2240, DD 10.97%, TT 72, Math 31.12
6) variant 5 + fractal dimension (BarsForFract indictor parameter) =6 (optimization 5-15 step 1) Profit 3226, DD 8.55%, TT 98, Math 32.92
7) Option 6 + SL (optimization 30-100 step 10) does not improve result, worse
8) Option 6 + TP 170 (optimization 50-200 step 10) Profit 3745, DD 3.89%, TT 98, Math 38.22
9) Option 6 + channel breakdown filter against direction Profit 3311, DD 10.60%, TT 78, Math 42.45

Files:
e-fxoe-shichannel_zzz.mq4  10 kb
 
ZZZEROXXX:

The bicycle is ready. Of course there is no alligator there, I wanted to see what I could squeeze out of one. Trade on channel breakdown.

What else we may try - a signal by channel slope, a signal on reversal of channel borders, combination of channels in a channel. The latter requires a proper source code.

p.s. In the tester, it is of course beautiful at an optimized period, but not so. Mixon is right, the analysis does not work. )))

 
ZZZEROXXX:

What else you can try - signal by channel slope, signal to bounce from channel borders, combination of channels in the channel. The latter requires a good source code.

p.s. In the tester it is of course beautiful on the optimized period, but not so. Mixon is right, the analysis does not work. )))

If I have watched it now I have noticed many unnecessary deals (I should not have deals inside the channel, only on the breakdowns). The deal to buy came earlier than the breakdown (this is dangerous!).

You have excluded the fractals and the alligator, while they are the main elements of the Williams trading system. The channel, as I said, only clearly indicates the breakdown, but the fractals and the alligator are the filters that guarantee the planned movement (in theory).

This channel indicator certainly changes the direction in the trend, but the deal opened even against the trend gives a minimum profit on the mean line of the channel.

P.S. The buy deal opened by the conditions of my technical task during the day had a loss, but did not reach SL and now I am in the black again.

 
I will test it on the tester when I arrive. I'll be sure to post the report and the results with the code.
 
Dizet_02:

Now I looked and noticed a lot of unnecessary deals (According to the principle there should not be deals inside the channel - only on breakdowns). A buy trade came before the breakdown (dangerous!).

In my EA?
 
ZZZEROXXX:
In my EA?
Yes. After the test , open the chart and choose a TF bigger than the test and see what the channel was and the trades within it.
 
I'm leaving Moscow today. I think that in my absence the branch will not go into a long box and will develop. On my arrival I'll be sure to report on the results of my trial.
 
Dizet_02:
Yes. After the test, open a chart and select a TF bigger than the test and see what the channel was and the trades within it.
Huh, why would you open on a TF larger than the one you tested? The channels on a larger TF will be totally different. If the "visualization" option is checked and the indicator has the same parameters as in the tester. This is the only way to assess the correctness of the algorithm, taking into account that channels are redrawn when new fractals appear.
 
Dizet_02:

You have excluded the fractals and the alligator, which are the main elements of the Williams trading system.

I do not want to repeat the Williams system, Roman has already done it. I did not use fractals and alligator because I wanted to see the efficiency of a pure channel-breaking system and then to compare it with other systems. I do not understand your scheme, I suppose - trade inside the channel on the bounce with trailing stops, enter after the fractal breakthrough and confirmation with alligator.
 
ZZZEROXXX:

Here's my lisp on the subject, for the tester. And some test results:

01/01/2010-01/01/2011
EUR/USD 1H Alpari
all ticks, simulation quality 90%
0.1 lot without MM

I don't understand how it's possible.
12345678910
New comment