This is a journal to explore some shareable ideas, and to attempt stuff I've not tried before. I don't promise anything useful.
- #1
- First Post: Oct 21, 2014 6:11pm Oct 21, 2014 6:11pm
- Joined Feb 2014 | Status: ember | 1,268 Posts
- #3
- Oct 21, 2014 6:23pm Oct 21, 2014 6:23pm
- Joined Jul 2009 | Status: Left Leg In, Right Leg Out! | 5,792 Posts
A pip is only worth it if you know how much you risked to earn it
- #4
- Oct 21, 2014 6:37pm Oct 21, 2014 6:37pm
- Joined Nov 2013 | Status: Member of the 5% club | 1,259 Posts
Yay! My very own guy started a thread! I'm so all over this one. 
- #7
- Oct 22, 2014 1:29am Oct 22, 2014 1:29am
- Joined Feb 2014 | Status: ember | 1,268 Posts
Have at it! It will be success filled I am sure. SUBSCRIBED
k
- #9
- Edited 5:40pm Oct 22, 2014 3:34pm | Edited 5:40pm
- Joined Feb 2014 | Status: ember | 1,268 Posts
Improved Bollinger Bands indicator.
In this comment PipMeUp discusses problems with the standard implementation of the Bands indicator.
His objections are:
1) The number of samples for the deviation should be typically much bigger than the one for the MA. This implementation allows this, as you can specify the numbers separately.
2) The lag of the simple moving average is half of its period. If we subtract the moving average at position [i] from the price at candle [i] to obtain the standard deviation in the next step, we are making an error, since the moving average because of the lag corresponds to the price half a period ago. So what one should be doing is to subtract the moving average at position [i] from the price at [i+MA_period/2]. This is fixed here.
The usual tricks (combining two timeframes to trade pullbacks):
In this comment PipMeUp discusses problems with the standard implementation of the Bands indicator.
His objections are:
1) The number of samples for the deviation should be typically much bigger than the one for the MA. This implementation allows this, as you can specify the numbers separately.
2) The lag of the simple moving average is half of its period. If we subtract the moving average at position [i] from the price at candle [i] to obtain the standard deviation in the next step, we are making an error, since the moving average because of the lag corresponds to the price half a period ago. So what one should be doing is to subtract the moving average at position [i] from the price at [i+MA_period/2]. This is fixed here.
The usual tricks (combining two timeframes to trade pullbacks):
- #10
- Oct 22, 2014 3:58pm Oct 22, 2014 3:58pm
- | Membership Revoked | Joined May 2012 | 1,218 Posts
Subscribe Sir.. 
MTH
MTH
- #11
- Oct 22, 2014 4:02pm Oct 22, 2014 4:02pm
- Joined Feb 2014 | Status: ember | 1,268 Posts
Subscribe Sir..
k
- #12
- Oct 22, 2014 4:11pm Oct 22, 2014 4:11pm
- Joined Jul 2009 | Status: Left Leg In, Right Leg Out! | 5,792 Posts
A pip is only worth it if you know how much you risked to earn it
- #13
- Oct 22, 2014 4:47pm Oct 22, 2014 4:47pm
- Joined Feb 2014 | Status: ember | 1,268 Posts
{quote} Hey K, Are you able to explain this a bit more - looks very interesting and I'd like to test it with my ZZBB method. Cheers V
As for the lag of the simple moving average (and other MAs) you can check here.
As for using it in the ZZBB, I think it could work reasonably well. Some stats should probably be done to check that the probabilities are OK...
k
- #14
- Oct 22, 2014 5:16pm Oct 22, 2014 5:16pm
- Joined Nov 2013 | Status: Member of the 5% club | 1,259 Posts
Improved Bollinger Bands indicator. In this comment PipMeUp discusses problems with the standard implementation of the Bands indicator. His objections are: 1) The number of samples for the deviation should be typically much bigger than the one for the MA. This implementation allows this, as you can specify the numbers separately. 2) The lag of the simple moving average is half of its period. If we subtract the moving average at position from the price at candle ...
- #15
- Oct 22, 2014 5:31pm Oct 22, 2014 5:31pm
- Joined Feb 2014 | Status: ember | 1,268 Posts
{quote} So I was wandering, I have an algorithm that relies heavily on Bill Williams Percent Range indicator. Is it possible to make any improvement on it, similar to what you've done with the BB here?
I think the oscillators should mostly be OK, but depending on the parameters, I could probably make it smoother, if this would help in your case. But this would probably not do such a big difference like for the BB.
k
- #16
- Oct 22, 2014 6:04pm Oct 22, 2014 6:04pm
- | Membership Revoked | Joined May 2012 | 1,218 Posts
{quote} Well, in whole honesty I only changed three-four lines of code, not really a massive effort involved. I think the oscillators should mostly be OK, but depending on the parameters, I could probably make it smoother, if this would help in your case. But this would probably not do such a big difference like for the BB. k
- #17
- Oct 22, 2014 6:10pm Oct 22, 2014 6:10pm
- Joined Feb 2014 | Status: ember | 1,268 Posts
{quote} Request your permission to add little spicy to 'her' for my price base chart environment (shifted renko)..
k
- #18
- Oct 22, 2014 6:17pm Oct 22, 2014 6:17pm
- | Membership Revoked | Joined May 2012 | 1,218 Posts
and the example result with default setting
EURUSD M15 + TZ h16 (4 + 4 hours)
EURUSD M11 shifted flip renko + TZ h16 (16 bricks) with 9.6 brick size (1440 bars M15 average size of EURUSD)
best regard
MTH
EURUSD M15 + TZ h16 (4 + 4 hours)
EURUSD M11 shifted flip renko + TZ h16 (16 bricks) with 9.6 brick size (1440 bars M15 average size of EURUSD)
best regard
MTH
- #19
- Oct 26, 2014 7:24am Oct 26, 2014 7:24am
- Joined Nov 2013 | Status: Member of the 5% club | 1,259 Posts
