Its all about "tests"

Yes, that is interesting.

Yet in my experience, when price "touches" the upper or lower channel line of a 2 standard deviation LRC, it almost never continues (on that same time frame). In my "walk forward" tests I have verified that price does indeed "retrace" about 95% of the time.

What does that tell you about "independence"?


Your comment is an important one, and it points out the need to refer to several time frames in order to make a decision to trade.

I guess I should have made this clear. You cannot trade from a single chart or time frame with this orientation.

Hope that helps

Steve
 
Quote from NickBarings:

I have to agree with fader.
More...

Certainly it is possible that this isn't the right approach for you.

One thing that can help you to sort out your difficulties is to do a little test

Simply put on four charts using 15 min, 5, 3 and 1 min bars or candles.

put in your LRC channels sizing them to 2 standard deviations and using 30 data points each.

Watch for a day or so and observe how the channels orient themselves in a cyclical way until at some point they "synchronize".

What you will find is that one does not need three digit accuracy to trade profitably. You only need rough approximation, and a couple of "landmarks" or "lines in the sand" to work from. Then it is a matter of your talent, your skill, your judgement. I say this because what it all comes down to in the end, is that you are looking at price wiggle at a specific point, and you are still faced with the same question, "do I enter here"? or "do I wait"?
 
Quote from steve46:

Yes, that is interesting.

Yet in my experience, when price "touches" the upper or lower channel line of a 2 standard deviation LRC, it almost never continues (on that same time frame). In my "walk forward" tests I have verified that price does indeed "retrace" about 95% of the time.
More...

well, since derivatives are a major part of today's markets and i'd guess that most derivative models pay special attention to 2+ std deviation moves due to leptokurtosis and "fat tails", it's not surprising that a 2 stdev trigger may carry special significance.

what you are saying is that you observe 100% of occurrences in your sample at a maximum of 2 stdev... rather than 95% - this would imply a non-normal distribution...

- what are the LRC settings you are using?
thanks for sharing, all the best.
 
Quote from fader:

well, since derivatives are a major part of today's markets and i'd guess that most derivative models pay special attention to 2+ std deviation moves due to leptokurtosis and "fat tails", it's not surprising that a 2 stdev trigger may carry special significance.

what you are saying is that you observe 100% of occurrences in your sample at a maximum of 2 stdev... rather than 95% - this would imply a non-normal distribution...

- what are the LRC settings you are using?
thanks for sharing, all the best.
More...

I have never used those words (100% of occurences).

Here (again) is what I did say.

"In my experience once price "touches" the upper or lower channel line of an LRC sized to 2 standard deviations, it ALMOST NEVER CONTINUES. In walk forward tests, I have verified that it does indeed "retrace" about 95% of the time. I mentioned that time frame is important, and I offered a suggestion for using multiple time frames to show how they orient to the distribution of price in a cyclical manner."

I hope this helps. Once again please note that this is simply a way to orient yourself to price movement. I use several additional tools to make my decisions.

I will have to stop for a while at this point. Talk to you folks later.

Steve
 
Quote from NickBarings:

Steve46,



It's OK, you don't need to contact your attorney !:cool:

I also use "test" set-up's like you.
More...

Well I thought I was being very circumspect in my comments.

Contacting attorneys is someone else's trademark I believe :)
 
You must log in or register to reply here.
Back
Top