New version 18 available now


Download software Tutorial videos
Subscription & data-feed pricing Class schedule


New account application Trading resources
Margin rates Stock & option commissions

Welcome Guest, please sign in to participate in a discussion. Search | Active Topics |

Clear Method-ish v17 Topic Rating:
Previous Topic · Next Topic Watch this topic · Print this topic ·
Al_Gorithm
Posted : Wednesday, August 09, 2017 11:32:00 AM
Registered User
Joined: 6/30/2017
Posts: 1,093
Hi Bruce,
 
Not sure how much of the Clear Method we can implement in v17. (It's easy peasy for StockFinder).
 
I'm assuming a variable lookback period for swings is not possible, unless you have some creative Ninja ideas on identifying the highest LOW during a variable-length upswing and the lowest HIGH during a variable-length downswing. (Note these may - or may not - be the same bar as the highest high upswing bar or the lowest low downswing bar.)
 
The basic idea borrows from a basic statistical concept that says if two distribtutions do not overlap, they can be considered different. Price bars on a larger timeframe are just simplified distributions of prices on a lower timeframe, so we can use the range of a bar to "guesstimate" when price has actually changed. (I know price action and statistics don't always play well together, but you gotta start somewhere.)
 
In a nutshell the Clear Method says when price clears below the highest low (HL), a new downswing begins. When price clears above the lowest high (LH), a new upswing begins.
 
If we can't finesse swings of variable length, then let's say for sake of example our lookback is 10 periods.
 
So, let's say ...
  • Change from upswing to downswing if current bar's high is lower than highest low in last 10 bars
  • Change from downswing to upswing if current bar's low is higher than lowest high in last 10 bars

Being able to "dot" the current bar if it's an upward clear bar or downward clear bar (below) would be awesome.

Not sure if we can dot the highest-low bars and the lowest-high bars, but if we could that we be even awesomer, even if a swing lasts a while and we wind up with multiple HL or LH dots during a swing. The last one would be the one folks care about. I can ignore the others.

 
 
 
 
 

Al_Gorithm
Posted : Wednesday, August 09, 2017 11:42:14 AM
Registered User
Joined: 6/30/2017
Posts: 1,093

I just noticed the PCFs you wrote for traderlady and bcochrane in the Formula for the High at the most recent Swing High topic.

Maybe something similar would work here to get around the 10-bar lookback. Those PCFs seem to go back 50 bars...

... except we'd be looking for the Highest Low of the most recent Swing High and the Lowest High of the most recent Swing Low.

Just a thought.

Bruce_L
Posted : Wednesday, August 09, 2017 12:41:12 PM


Worden Trainer

Joined: 10/7/2004
Posts: 63,198

Well I don't really have a way to color the price bars (but we could use dots I guess). More importantly, resetting the direction of the trend and then starting to count again requires either storing a variable or some sort of recursive function (and recursive functions make for very long PCFs).

If we don't reset, we can plot something like the following using a Custom PCF Indicator and the Histogram Plot Style. The magnitude is how many bars back it had to look for the bar to be clear of other bars.

So a large magnitude can represent a long trend but seems more likely to represent weak performance of that particular bar versus the trend. I almost wonder if putting 100 / in front of the fomula would make the results more intuitive (although that is a bit simplistic and probably not the best way to "reverse" the magnitude of the bar).

IIF(L > H1, 1, IIF(H < L1, -1, IIF(L > MINH2.1, 2, IIF(H < MAXL2.1, -2, IIF(L > MINH3.1, 3, IIF(H < MAXL3.1, -3, IIF(L > MINH4.1, 4, IIF(H < MAXL4.1, -4, IIF(L > MINH5.1, 5, IIF(H < MAXL5.1, -5, IIF(L > MINH6.1, 6, IIF(H < MAXL6.1, -6, IIF(L > MINH7.1, 7, IIF(H < MAXL7.1, -7, IIF(L > MINH8.1, 8, IIF(H < MAXL8.1, -8, IIF(L > MINH9.1, 9, IIF(H < MAXL9.1, -9, IIF(L > MINH10.1, 10, IIF(H < MAXL10.1, -10, IIF(L > MINH11.1, 11, IIF(H < MAXL11.1, -11, IIF(L > MINH12.1, 12, IIF(H < MAXL12.1, -12, IIF(L > MINH13.1, 13, IIF(H < MAXL13.1, -13, IIF(L > MINH14.1, 14, IIF(H < MAXL14.1, -14, IIF(L > MINH15.1, 15, IIF(H < MAXL15.1, -15, IIF(L > MINH16.1, 16, IIF(H < MAXL16.1, -16, IIF(L > MINH17.1, 17, IIF(H < MAXL17.1, -17, IIF(L > MINH18.1, 18, IIF(H < MAXL18.1, -18, IIF(L > MINH19.1, 19, IIF(H < MAXL19.1, -19, IIF(L > MINH20.1, 20, IIF(H < MAXL20.1, -20, IIF(L > MINH21.1, 21, IIF(H < MAXL21.1, -21, IIF(L > MINH22.1, 22, IIF(H < MAXL22.1, -22, IIF(L > MINH23.1, 23, IIF(H < MAXL23.1, -23, IIF(L > MINH24.1, 24, IIF(H < MAXL24.1, -24, IIF(L > MINH25.1, 25, IIF(H < MAXL25.1, -25, IIF(L > MINH26.1, 26, IIF(H < MAXL26.1, -26, IIF(L > MINH27.1, 27, IIF(H < MAXL27.1, -27, IIF(L > MINH28.1, 28, IIF(H < MAXL28.1, -28, IIF(L > MINH29.1, 29, IIF(H < MAXL29.1, -29, IIF(L > MINH30.1, 30, IIF(H < MAXL30.1, -30, IIF(L > MINH31.1, 31, IIF(H < MAXL31.1, -31, IIF(L > MINH32.1, 32, IIF(H < MAXL32.1, -32, IIF(L > MINH33.1, 33, IIF(H < MAXL33.1, -33, IIF(L > MINH34.1, 34, IIF(H < MAXL34.1, -34, IIF(L > MINH35.1, 35, IIF(H < MAXL35.1, -35, IIF(L > MINH36.1, 36, IIF(H < MAXL36.1, -36, IIF(L > MINH37.1, 37, IIF(H < MAXL37.1, -37, IIF(L > MINH38.1, 38, IIF(H < MAXL38.1, -38, IIF(L > MINH39.1, 39, IIF(H < MAXL39.1, -39, IIF(L > MINH40.1, 40, IIF(H < MAXL40.1, -40, IIF(L > MINH41.1, 41, IIF(H < MAXL41.1, -41, IIF(L > MINH42.1, 42, IIF(H < MAXL42.1, -42, IIF(L > MINH43.1, 43, IIF(H < MAXL43.1, -43, IIF(L > MINH44.1, 44, IIF(H < MAXL44.1, -44, IIF(L > MINH45.1, 45, IIF(H < MAXL45.1, -45, IIF(L > MINH46.1, 46, IIF(H < MAXL46.1, -46, IIF(L > MINH47.1, 47, IIF(H < MAXL47.1, -47, IIF(L > MINH48.1, 48, IIF(H < MAXL48.1, -48, IIF(L > MINH49.1, 49, IIF(H < MAXL49.1, -49, IIF(L > MINH50.1, 50, IIF(H < MAXL50.1, -50, 0))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))

But I think it can identify trends which aren't there because it doesn't reset and can look back too far as a result.



-Bruce
Personal Criteria Formulas
TC2000 Support Articles
Al_Gorithm
Posted : Wednesday, August 09, 2017 1:06:25 PM
Registered User
Joined: 6/30/2017
Posts: 1,093

Yeah, using dots until v18 allows us to paint price bars is fine. I know the devs are working hard on that as we speak ;)

And resetting swing direction is always the sticky bit without being able to store variables and do some looping. I've never done much with recursion, mostly because I've never had to, and because thinking recursively makes my head hurt. :)

Your solution seems really cool. Would never had thought of plotting a histogram to indicate the number of bars from the HL or LH to the clear bar. Very cool!!

I was the one who actually turned Julia onto the Clear Method which she highlighted in a 2011 webinar with Michael. I gave Ron Black a heads-up before the webinar, and he attended. Later he emailed me suggesting using a 5-bar lookback which he observed worked pretty well. But I'm thinking we'd need more like 10 bars since some of the swing examples in his own articles exceed 5 bars. 

I'll start with 10 bars as the PCF stands now and see how it goes. Can always adjust later. The Clear Method ain't the holy grail but it does provide a specific unambiguous objective definition of swing reversals, and anything that helps in v17 works for me. I really don't want to go back to StockFinder even though the programmer in me misses writing code. Always wanted to combine Clear Method swings and basic Dow theory (higher highs, higher lows, etc...) to combine swings into trends. I know v17 can't handle that. LOL.

Thanks, Bruce!

Al_Gorithm
Posted : Wednesday, August 09, 2017 1:08:29 PM
Registered User
Joined: 6/30/2017
Posts: 1,093

Whoops, looks like you edited the post while I was responding. Reading latest version now.

Bruce_L
Posted : Wednesday, August 09, 2017 1:09:13 PM


Worden Trainer

Joined: 10/7/2004
Posts: 63,198

I think I modified my post after you started replying. The new longer formula should be better (I think the short version actually had a logic error).



-Bruce
Personal Criteria Formulas
TC2000 Support Articles
Al_Gorithm
Posted : Wednesday, August 09, 2017 1:16:45 PM
Registered User
Joined: 6/30/2017
Posts: 1,093

That latest PCF looks a lot longer than the original. :)

I'll wait until I hear back before I start noodling around. Looking back too far may indeed be an issue. And you keep mentioning "trends" while I refer to "swings" ... just want to make sure we're on the same page ... to me a trend consists of multiple swings, up and down, as long as the trend continues in the same direction. 

The Clear Method isn't trying to identify trend reversals, just swing reversals. Hope that makes sense.

Al_Gorithm
Posted : Wednesday, August 09, 2017 1:22:35 PM
Registered User
Joined: 6/30/2017
Posts: 1,093

D*mn it, I'm still not typing fast enough. :)

I'll try your new longer formula. Your concept is very Guru-Jedi-Ninja-RockStar-Genius like.

Bruce, you're awesome. Take a coffee break - you've earned it.

 

 

Bruce_L
Posted : Wednesday, August 09, 2017 1:25:25 PM


Worden Trainer

Joined: 10/7/2004
Posts: 63,198

The clear method really only gives you information about the strength of a single bar versus previous bars.

So if I were going to use either swing or trend terminology, I would think a clear bar might indicate a swing, but not a trend.

That said, if we are counting the number of bars required for the bar to be clear, we could be getting an indication of the strength of that bar vs the trend. The bigger the count, the weaker that bar is versus any existing trend (although it could also signal the start of a breakout from a consolidation).

And a long series of low magnitude counts in the same direction would probably constitute an actual trend. When the magnitude starts to increase, the individual bars are getting weaker versus the trend (and thus, the trend itself would be weakening or maybe even disappearing altogether).



-Bruce
Personal Criteria Formulas
TC2000 Support Articles
Al_Gorithm
Posted : Wednesday, August 09, 2017 1:39:01 PM
Registered User
Joined: 6/30/2017
Posts: 1,093

I agree with your second sentence: "... a clear bar might indicate a swing."

The rest I'm a little fuzzy on, but we may both be saying the same thing differently so no worries. Going to sleep on it and reread your reply. Let me kick the tires. I never imagined a histogram with counts so I'm not sure how that'll play out. I wasn't really thinking about the strength of a bar or a trend - I was just thinking of identifying swing reversal points. The histogram is a bonus. :)

Actually, I just got the meaning of your last sentence. There's hope for me yet. LOL.

Thanks, Bruce!

Al_Gorithm
Posted : Wednesday, August 09, 2017 1:53:56 PM
Registered User
Joined: 6/30/2017
Posts: 1,093

Rewatching that webinar now. Really miss Julia. <sigh>

If anyone is interested in the Clear Method, I've got a link to the orignal article above. Ron Black also wrote a follow-up article a few months later that you can look up.

Julia also does a great job explaining the concept in her webinar. Worth watching.

And Bruce's v17 PCF will soon be world famous. :)

Al_Gorithm
Posted : Wednesday, August 09, 2017 2:33:24 PM
Registered User
Joined: 6/30/2017
Posts: 1,093

Bruce,

I know Julia has moved on from Worden, but around the 43-minute mark of her webinar she admitted both her and Michael couldn't come up with a "Clear Method" PCF ... she offered to buy dinner for anyone who came up with a viable PCF. Of course the time variability of the swings stumped her.

You might want to take her up on it. :)

diceman
Posted : Thursday, August 10, 2017 10:09:50 AM
Registered User
Joined: 1/28/2005
Posts: 6,048

Are the zero crossings the CLEAR buy, sell signals?

 

 

 

Thanks

 

Al_Gorithm
Posted : Thursday, August 10, 2017 10:35:28 AM
Registered User
Joined: 6/30/2017
Posts: 1,093

diceman,

I turned off my computer last night and went for a looong walk in some gorgeous weather, so I haven't tried Bruce's PCF yet.

For me the Clear Method is just an objective technique for identifying when price changes direction (i.e., price "swings"). The original author really doesn't talk about "signals" or "systems" ... the indicator (a "clear bar") tells you when price has started a new upswing or a new downswing based on the statistical concept of overlapping distributions, which are just considered "noise". When one distribution (bar) "clears" another, you can say price has changed.

I'd check out the PDF (link above). It's a short quick read, clear (pun intended) and concise. A few minutes tops. If you're going to use Bruce's Jedi Master Histogram you should probably understand the original concept first. If you have time and are interested Julia's webinar (link above) is must see TV. 

Having said all that, Bruce took the idea to a whole new level with a histogram that counts back from the current "clear bar" to the latest highest-low or lowest-high. I still haven't wrapped my head around how this translates to "strength" of bars or trends, but I have to play around with it and kick the tires.

So, to make a long story short (LOL!) I'd recommend reading the 4-page PDF. If you like that check out Julia's webinar. And if you like that, check out Bruce's histogram on a bunch of charts. I really won't understand what the zero line crosses signify, if anything, until I check it out myself.

Given Bruce's mad skills I'm guessing he's onto something - I just don't know what yet. :)

diceman
Posted : Thursday, August 10, 2017 11:41:40 AM
Registered User
Joined: 1/28/2005
Posts: 6,048

i know what it is, I just wanted to avoid going through the PCF.

 

What Bruce created looks better vs the actual Clear Method.

 

 

 

 

Thanks

 

Al_Gorithm
Posted : Thursday, August 10, 2017 11:59:27 AM
Registered User
Joined: 6/30/2017
Posts: 1,093

Got it ... I avoid going through Bruce's PCFs at all costs. They're genius but the syntax gives me a headache. LOL.

Looking forward to checking it out.

Thanks Bruce! Thanks diceman.

dhirenshah
Posted : Sunday, June 03, 2018 2:31:47 PM
Registered User
Joined: 3/10/2012
Posts: 352

Bruce 

I was going through this article above there is some serious swing idea to entertain here... 

your formula that u made , can we overlay it on the price with two different colours  in V18  that you did as histogram??

 

And one non-related query in V18  do we have the facility to show area filled "only when WRSI14 crossed above 70 or below 30 levels ? or can that be customized to overlay when it happens ?? same with Stochastics 12-3-3 can area filled be shown when stochastics goes below 20 or goes above 80 ??

Thanks for the Post Al gorith, its some interesting idea...

 

Regards

Dhiren

 

Bruce_L
Posted : Monday, June 04, 2018 11:12:43 AM


Worden Trainer

Joined: 10/7/2004
Posts: 63,198

I have sent you a copy of the most recent version I have made of the Clear Method-ish indicators via TCMail.

There really isn't a way to conditionally fill in TC2000.



-Bruce
Personal Criteria Formulas
TC2000 Support Articles
dhirenshah
Posted : Monday, June 04, 2018 12:26:46 PM
Registered User
Joined: 3/10/2012
Posts: 352

Bruce this is Amazing , TC is going to the whole new level :) Brilliant work

thanks a lot 

Although one question in context to the Magnitude part.. 

if you see those high magnitude stand alone bars in histogram in any chart and your interpretation that when those individual bars increase in magnitude the bars are getting weaker so ... i m wondering is there a way to make a Parabolic SAR to show the increase of magnitude below and above the price ?? and mark the climax weak bar (with most increment) with another Dot on price or colour change?

 

Bruce_L
Posted : Monday, June 04, 2018 2:40:23 PM


Worden Trainer

Joined: 10/7/2004
Posts: 63,198

I have sent you a copy of the most recent version I have made of the Clear Method-ish indicators via TCMail.

There are dots and the dots change color, but how these dots should be interpreted is entirely up to you (and I am not sure Parabolic SAR would be a good guide).

The yellow dots represent the most recent value cleared by the most recent bar in either direction.

The green dots represent the most recent cleared value below the most recent bar (unless that value currently has a yellow dot).

The red dots represent the most recent cleared value above the most recent bar (unless that value currently has a yellow dot).



-Bruce
Personal Criteria Formulas
TC2000 Support Articles
dhirenshah
Posted : Monday, June 04, 2018 3:23:06 PM
Registered User
Joined: 3/10/2012
Posts: 352

Bruce this is fantabulous... fantastic and fabulous 

Thank you very much for taking time to build it you orchestrate the chart in such a wonderful manner that only thing left is to make money :)) 

Cheers

Dhiren

Al_Gorithm
Posted : Monday, June 04, 2018 5:13:36 PM
Registered User
Joined: 6/30/2017
Posts: 1,093

Bruce, can you send me your latest and greatist version?

Al_Gorithm
Posted : Tuesday, June 05, 2018 11:13:47 AM
Registered User
Joined: 6/30/2017
Posts: 1,093

Got it. Thanks, Bruce!

raynd
Posted : Tuesday, June 05, 2018 2:16:00 PM
Gold Customer Gold Customer

Joined: 5/11/2013
Posts: 14
Bruce, pls send me copyof clear method indicator havesomething similar with my broker platform want to compare. Tks
dhirenshah
Posted : Tuesday, June 05, 2018 9:56:59 PM
Registered User
Joined: 3/10/2012
Posts: 352

Bruce 

after testing and making good day trades . i just need some additional pcf from you.. in this method for STRENGHT OF MAGNITUDE...if its possible please

In your reply above Wednesday, August 09, 2017 1:25:25 PM

you said : bigger the count, the weaker that bar is versus any existing trend (although it could also signal the start of a breakout from a consolidation)

So my question is taking example of AMZN"

magnitude on daily chart for MAY 23,2018 was the highest BELOW zero so consider that as a " CLIMAX BAR" and next day stock MOVES HIGHER 

Another example BBY " on daily chart had a climax bar magnitude ABOVE ZERO MAY 22,2018 and next few bars were DOWNSIDE... 

Now my question is if i want to only scan that climax bar " up or down(which is the histogram biggest bar out of ordinary) do you think you can PCF the following

1) tallest HISTOGRAM BAR that can be plotted as pcf spike on price chart where the MAGNITUDE OF THE RED BARS IN THE COUNT of bars BELOW ZERO WAS THE HIGHEST before the GREEN bar histogram started "  and ENDED IN THE PAST please see attached AMZN image 

https://www.dhirenshah.com/clear-bar-1/

2) HISTOGRAM BAR that can be plotted as dot or pcf spike on price chart where the MAGNITUDE of the GREEN  BARS IN THE COUNT of bars ABOVE ZERO was the HIGHEST before the new RED BAR histogram started AND ENDED IN PAST please see attached link image of BBY"

https://www.dhirenshah.com/clear-2/

 

I want to sort and scan them 

thanks again 

Bruce_L
Posted : Wednesday, June 06, 2018 9:56:21 AM


Worden Trainer

Joined: 10/7/2004
Posts: 63,198

I cannot think of a way to write such a formula which would be fast and short enough to be practical or post in the forums.



-Bruce
Personal Criteria Formulas
TC2000 Support Articles
dhirenshah
Posted : Wednesday, June 06, 2018 10:16:16 AM
Registered User
Joined: 3/10/2012
Posts: 352

No worries Bruce then i guess will stick with best scan of eyeballing them :)) 

Cheers

steple99
Posted : Wednesday, June 20, 2018 1:37:56 PM
Gold Customer Gold Customer

Joined: 12/6/2004
Posts: 3

Hello Bruce,

I would appreciate it if you could send me the latest PCF for the Clear Method.

Thanks!

bcochrane
Posted : Wednesday, June 20, 2018 4:27:10 PM
Registered User
Joined: 9/17/2010
Posts: 401

Bruce, if I may have a copy as well.

Thanks

bcochrane
Posted : Thursday, June 21, 2018 8:39:29 AM
Registered User
Joined: 9/17/2010
Posts: 401

Got it.

Thanks Bruce

bostonbombay
Posted : Wednesday, July 04, 2018 9:13:18 AM
Gold Customer Gold Customer

Joined: 6/23/2015
Posts: 12

Hi Bruce,

Can I get a copy of the same.

Regards,

MK

Partha223
Posted : Friday, July 06, 2018 10:49:25 PM
Registered User
Joined: 2/13/2018
Posts: 24

Hi Bruce,

 

Appreciate if you can send me a copy of this.

Thanks in advance

 

bostonbombay
Posted : Sunday, July 08, 2018 9:18:30 PM
Gold Customer Gold Customer

Joined: 6/23/2015
Posts: 12

QUOTE (Bruce_L)

I have sent you a copy of the most recent version I have made of the Clear Method-ish indicators via TCMail.

 

Thanks Bruce. Appreciate it.

 

Regards,

MK

Bruce_L
Posted : Monday, July 09, 2018 10:53:35 AM


Worden Trainer

Joined: 10/7/2004
Posts: 63,198

I have sent you a copy of the most recent version I have made of the Clear Method-ish indicators via TCMail.



-Bruce
Personal Criteria Formulas
TC2000 Support Articles
Users browsing this topic
Guest-1

Forum Jump
You cannot post new topics in this forum.
You cannot reply to topics in this forum.
You cannot delete your posts in this forum.
You cannot edit your posts in this forum.
You cannot create polls in this forum.
You cannot vote in polls in this forum.