thanks acetrader
- #13,309
- Jun 27, 2018 11:02am Jun 27, 2018 11:02am
- | Membership Revoked | Joined Jun 2007 | 3,782 Posts
- #13,310
- Jun 27, 2018 11:03am Jun 27, 2018 11:03am
- | Membership Revoked | Joined Jun 2007 | 3,782 Posts
- #13,312
- Sep 18, 2018 2:33pm Sep 18, 2018 2:33pm
- Joined Aug 2015 | Status: Coding Magician | 2,676 Posts
Vucking good EA coder... https://t.ly/AZjRM
- #13,314
- Dec 20, 2018 9:01pm Dec 20, 2018 9:01pm
- Joined Nov 2006 | Status: Trader | 4,417 Posts
to trade and code, keep both simple... no call to impress....h
- #13,315
- Edited 5:09am Dec 21, 2018 4:03am | Edited 5:09am
{quote} //------ hey anti...... yes i remember this thread.... and what it later turned into..... when there becomes alot of squabblin, i usually move on..... as far as similarity stuff, that is very real..... more to do with the math than anything else..... anytime you increase the timeframe by a factor and decrease the indicators setting by the same factor, the indicators will be similar..... such as the 20 ma on a 60 minute chart is similar to a 10 ma on a 120 minute chart..... you doubled the time and cut the periods in half.... any trend change...
This behavior can be prevented when using an MTF version of the ZZ that shows all legs (the ZZ uses highs and lows instead of close prices). But then there are (almost) no dissimilarities left.
Another problem is the "Major Cycle Length" setting (or in the ZZ indicator the "Backstep"). Even if you use a non-repaint version of the CI (like the Symphonie Extreme Indicator), it forces the indicator to delete repainted signals. Thus, you can only see (with a non-repaint indicator) where signals appeared if you set it to 0. However, some guys @Similarity thread (like juhanimi or Madmoney) seem to have traded well with the CI, regardless its repainting behavior.
Now let's force the ZZ to show all repainted signals:
How would you trade it?
Or how would you trade different ZZs with different settings within a TF? The general idea should be clear: Since the ZZ (when a high period cycle length is used) prolongs a leg quite often, you could wait to see a contrary signal in the shorter period setted ZZ and use it to enter the trend (shown by the longer period ZZ). But this fails/eats up all profits when the trend comes to an end ...
- #13,316
- Dec 21, 2018 6:53am Dec 21, 2018 6:53am
- Joined Nov 2006 | Status: Trader | 4,417 Posts
to trade and code, keep both simple... no call to impress....h
- #13,319
- Edited 3:05pm Dec 30, 2018 10:01am | Edited 3:05pm
Here are two interpretations of the similarity idea. It is crucial to understand that a new trend can only start if there were two signals that agreed. Since I used my ZZ indicator with 2 different Depth settings (24 and 96), a new trend can only start if there is either an up or a down spike in both indicator windows at the same time. The ZZ indicator can repaint a lot. We will usually see a multitude of signals in both windows that spike in the same direction and occur subsequently until a trend is exhausted and a new trend starts. Thus, the best choice for the confirmation of a new trend is to wait for two spikes in the opposite direction than the last spikes that agreed.
Since we expect to see a higher number of repaints/trend continuations, we could trade this in two ways (both require the determination of the main trend as shown above and are described for up trends only):
Since we expect to see a higher number of repaints/trend continuations, we could trade this in two ways (both require the determination of the main trend as shown above and are described for up trends only):
- Wait for an up signal in the upper ZZ window without an up signal is shown in the lower ZZ window. Then we could target the high of the candle where the signal occured and trade long with the open of the next candle (or any retracement lower), expecting a higher high to be made (blue/red dots for possible longs/shorts).
- Wait for a down signal in the upper ZZ window. This may signal a retracement within the up trend and thus could be a good price point to go long (blue/red squares for possible longs/shorts).
However, both attempts will fail when the trend comes to an end. And in some cases you'll also encounter a bigger DD.
