Analysis and Trading based on Envelopes, Waves, Cycles | Page 592 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Planetary cycles, astrological theory of cycles of the markets 510 replies

Forecast Forex using Numeric Cycles and Astrology Cycles 38 replies

Planetary Cycles... Plotting Cycles into the Future 0 replies

Market Waves are NOT Cycles 8 replies

Waves/Trends/Cycles 0 replies

Attachments: Analysis and Trading based on Envelopes, Waves, Cycles
Exit Attachments

Analysis and Trading based on Envelopes, Waves, Cycles

  • #11,821
  • Edited 3:00pm Dec 11, 2023 2:23pm | Edited 3:00pm
Quoting OAPDave
{quote} Terry Jumping Jack Flash That's amazing, just as I remember it Murrey licensed his software, hence you had to (hope) he would send you a new password every October, alas after a couple of years he just disappeared Obviously no password caused the software to die Your effort is bang on, in fact brilliant, now what you need to watch is how and were price enters the next TradeFrame Also his, by pressing any of the 4 8 16 32 64 128 265 button you would see the relevant TradeFrame and that was a bit of a problem, it just took an age Also you...
Dave I do not trust Murrey.

I do not trust Murrey for the details he wrote. Nothing is proved and demonstrated
I do no trust the Software.

Have you ever seen a User Guide for this software?
Have you ever seen a Trading Guide ? How to trade with Murrey's software ?

The advantage I have on some people is that I was involved in the Murrey's Methodology as soon as 1998 or 1999. I bought the software but could not do a lot with it because of its lack of versatility. I gathered all what I could about the Murrey's environment at this time.

This is when I used Murrey's ideas (explained by Kruzel) with Excel.

Moreover I was interested by a Murrey's group which had similar questions on the method. The founder of the group Brian McMullan developed a software or an add-in (I do not remember).

This was the time with many discussions. Let's say that most questions or interrogations you , Terry and other people here have were discussed or adressed as soon as 2000.
  • #11,822
  • Dec 11, 2023 2:29pm Dec 11, 2023 2:29pm
Brian Mc Mullan

MM Time Frame: In Menu item Options, there is a new item called
"Frame Length" with two sub items called "Natural" and
"Natural * 1.5 + 1"
Natural - a 32 frame will use 32 bars
Natural*1.5+1 - a 32 frame will use 32*1.5+1 = 49 bars

Why??? Those who have studied Murrey's program say "Natural * 1.5 + 1" is what Murrey uses in his program, so I offer it as an option.

Brian
At the seminar in Beverly Hills Murrey talked about how he determined the high and low of the frame .......he said for a 16 day frame he used the range of the last 32 days...and if prices closed past 10/8 or -2/8 he would adjust the frame. Murrey also mentioned something about the prices going past 8/8 and 0/8 for several days to adjust the frame. Steve ...the computer guy was in the front row. I know he is with this group or the other. Also try agent moulder he's good at details. Ron

Yes... Murrey says 2X. That's not what the s/w does though. I don't know if Murrey is intentionally misleading the audience or if Werling has done something different than Murrey thinks he has.

Bruce


I have the feeling that the true answer lies somewhere in between. I think that Murrey's software looks back up to two times the time period chosen and select only the first significant high and low that matter. If there is another significant high and low before the one that it picked, still within the two times the period that was selected, it ignores it.


If I had more details I would not say "I have the feeling", but I would say it is. Anyway, my feeling is:

- say you set the frame to 16. The software should look back 32 and see what significant highs and lows are present. Based on an algorithm it would then determine what is "significant", and then choose the closest "significant" to the present and drop the one before it, even if the one before is more "significant". Murrey keeps saying that the last high and low are the ones that count. But in fact they have to be significant highs and lows. It does not really matter where they are located, as long as they are the LAST significant high and low.

Brian's software now looks at the high and low within the designated period and adds lines based on the highest high and lowest low present in that period. I don't think that is the correct way of doing it. Adding a range finder and then computing the "significant" highs and lows, then choosing the last "significant" high and low are the way to go, IMHO. One should be easily able to compare this thought by comparing the daily charts from Murrey's software and the daily charts with Brian's software. You will probably have to go through fifty or more charts, and see what settings in Brian's software will equate with Murrey's daily chart set, say, at 16 periods. Do not change the periods in Murrey, but change them in Brian's to obtain the same results as in Murrey's. Use charts that have good price excursion.

I am planning to do just this as soon as Murrey sends me the latest software that will read data after last December. I hope this explains my "feelings". Cheers. Bijan
  • #11,823
  • Dec 11, 2023 2:34pm Dec 11, 2023 2:34pm
In recent versions I added the 1.5x & 2.0x multipliers (eg if using a 64 frame, then 2.0x would use the extreme high/low in the last 128 days).

A recent post by Murrey lends credence to using ONLY "natural" or 1x settings though...

The public post was dated 2/1/2k and titled "MMTS: OEX Support off current lows:
Downside almost over: Reverse soon" Murrey shows the OEX in a 64 frame (8 days per 1/8th time interval).

To make XChartMM "match" his graph, the 64 frame setting is used along with the 1x multiplier - eg, NO multiplier.


I've heard this from Murrey before. The issue is that his s/w doesn't work that way. Remember that Kristof Werling is the one who created the code and what he's done may not match Murrey's desires. What the s/w does is 1.5*N+1, even in 64-bar frames.

That aside, you have all the bases covered, so it's not a problem.

Bruce


Wait - what I'm saying is that if Murrey's prog really used 1.5n+1, then the output of his recent OEX graph *should* be different, because the price extremes were larger at 1.5n than at 1.0n.

In other words, Murrey's OEX graph contradicts the 1.5n+1 scenario for all intents and purposes
  • #11,824
  • Dec 11, 2023 2:38pm Dec 11, 2023 2:38pm
Guys,

I think this may prove my point that regardless what time frame you put yourself in, the software looks for the most recent "significant" high and low to calculate the frame size.
Another chart which implies that 1.5x is the one to use as a general rule (this one is frame size 16).

On 1x, the high and low in 16 days do not compute the frame to match THMurrey's output.

I agree, Brian. The only thing I've seen besides 1.5X+1 is a shorter display that seems to be reacting to a rule they have about significant lows/highs. The ABTXQ chart does that after a whipsaw in the rally that precedes the displayed portion of the chart. I have to assume the Murrey s/w looks for reversals off intermediate highs and if they're greater than X%, he says it's significant. I don't recall reading about that in the book, but I think it's in the s/w.

Bruce

Murrey's EOD s/w currently uses a greater number of bars than the frame size to test for significant highs/lows. The number of bars tested is 1.5*N+1 where N is the frame size. So for a 16-bar frame, the number of bars tested for significant high/low is1.5*16+1 = 25 bars. So a 16-bar frame, if allowed to display unassisted, will usually show 25 bars of data, the 32-bar frame will display 49 bars and the 64-bar frame will display 97 bars.

Murrey's s/w also has a whipsaw detector (of sorts) that tests to see if a whipsaw is sufficient to constitute a significant high/low. If the detector gets activated, the number of bars displayed will be less than 1.5*N+1.

Bruce
  • #11,825
  • Dec 11, 2023 2:40pm Dec 11, 2023 2:40pm
Pretty simple really

With the frame settings of 4, 8, 16, 32 & 64; the program finds the highest high/lowest low in the last X bars, where X is:

on "1x", or natural, frame setting x 1, eg 64 is bars
on "1.5x", it is frame setting x 1.5, eg 64 is 64 * 1.5, or 96 bars
on "2x", it is frame setting x 2.0, eg 64 is 64 * 2, or 128 bars

I personally use the 2x setting, this seems to match Murrey's program output in the majority of cases. Brian
  • #11,826
  • Dec 11, 2023 2:53pm Dec 11, 2023 2:53pm
Quoting OAPDave
{quote} I have not bumped into that Did you ever use it in the early days ?
What Brian McMullan wrote about Kruzel's notes and Kruzel's DOS Program :

Dick, I had direct interchanges with Tim Kruzel some 18 months to 2 years ago and noted or recall the following:

1) his widely disseminated notes were a rough draft; Tim Kruzel never intended for them to "get out" - that just kind of happened after he posted his e-mail to THMurrey.

2) The process defined in Tim Kruzel's notes was a pretty good start; but is by no means the "right way" and (as Kruzel says) there are fundamental flaws.

3) Tim Kruzel AMM.EXE program uses later revision(s) of the algorithm. His metastock code is somewhere between the "notes" and "AMM".

4) Tim Kruzel's work past AMM.EXE; never made public or put into a program except for his own use.

5) Tim Kruzel doesn't trade exclusively on Murrey Math.

6) XChartMM is not "Kruzel's notes" applied graphically; I too started with his notes but quickly found they were not right (as Tim confirmed). Then I started to work, got a clue or two from Tim, and some time later, well, here we are.

But, only Murrey knows Murrey's actual algorithm! Tim doesn't know it, I don't know it.


7) There are a lot of "MM add-ons" in use for tradestation, metastock, this & that; most of them are 100% based on Kruzel's rough draft notes (which are not right as we already discussed and Kruzel confirmed).

Hence, there are a lot of frustrated people out there who say "Murrey's stuff doesn't work".

It's not "Murrey's stuff" unless you have Murrey's actual stuff!! Go to his classes and use his software to get the "real thing"!

This isn't an advert for THM either! It is "what is", plain and simple.
  • #11,827
  • Dec 11, 2023 4:19pm Dec 11, 2023 4:19pm
my last short
Attached Image (click to enlarge)
Click to Enlarge Name: screenshot.png Size: 69 KB
  • #11,828
  • Dec 11, 2023 5:14pm Dec 11, 2023 5:14pm
Quoting parisboy
my last short {image}
That's a beauty!
  • #11,829
  • Dec 11, 2023 5:31pm Dec 11, 2023 5:31pm
Thanks Alain. Based on that info I will leave the calculation of high/low on automatic. At least for now anyway.

In terms of drawing vertical time lines, please help out MM "old timers"

It seems from the image of the software that the vertical lines were spaced at 4,8,16,32,64,128 or 256 bars.
1. Was the placement of the selected spacing adjustable so that you could match it to the price movement?
2. Have we decided where the 8 block square should originate from. Perhaps the low/high used in the calculation for the price line?
  • #11,830
  • Dec 11, 2023 5:50pm Dec 11, 2023 5:50pm
Terry, brief answer tonight: from TM's original course, see point 6 for the frame start date:
Attached Image (click to enlarge)
Click to Enlarge Name: MM-startdate.png Size: 392 KB
  • #11,831
  • Dec 11, 2023 6:08pm Dec 11, 2023 6:08pm
the powerpoint from where the above chart is extracted
Attached File(s)
File Type: pdf T. Henning Murrey - Murray Math Trading System Manual.pdf   7.5 MB | 725 downloads
  • #11,832
  • Dec 11, 2023 6:08pm Dec 11, 2023 6:08pm
Quoting GSQ120
Terry, brief answer tonight: from TM's original course, see point 6 for the frame start date: {image}
Thanks GSQ.
First frost each month, I do remember reading that, and also that Murrey would not divulge where that first frost was located. Based on that it also seems that Murrey only reset the trading frame starting point annually.

I may be wrong, but I'm guessing the starting annual starting point didn't matter, as long as all users of the software were using the same date/time.
  • #11,833
  • Dec 11, 2023 6:12pm Dec 11, 2023 6:12pm
Quoting tmpascoe
Thanks Alain. Based on that info I will leave the calculation of high/low on automatic. At least for now anyway. In terms of drawing vertical time lines, please help out MM "old timers" It seems from the image of the software that the vertical lines were spaced at 4,8,16,32,64,128 or 256 bars. 1. Was the placement of the selected spacing adjustable so that you could match it to the price movement? 2. Have we decided where the 8 block square should originate from. Perhaps the low/high used in the calculation for the price line?
Terry, I had the 1999 / 2000 version of the software . Nothing was adjustable . No user guide.

It is why Iused Excel with Daily Data at the time
  • #11,834
  • Dec 11, 2023 6:20pm Dec 11, 2023 6:20pm
If you are concerned about price levels they are unaffected by the start date. As for the time lines, you might be interested in the following which are portions of two messages from Bonnie Hill (her website is http://home.flash.net/~blhill/pages/murrey.html).

Bill

I need to give a little history of "Murrey Math".

"Murrey Math" is based on a course written by W D Gann. When Murrey put his mother into a rest home, he found the course when he cleaned out her house. He used this material as the basis for "Murrey
Math". In the early days, he gave Gann credit for the material.

As the years passed, Murrey has "forgotten" where the ideas came from :-) There are several ideas in the course which Murrey did NOT understand correctly. That is why some of the techniques seem great and some seem a little "strange".

The starting point for the 4th quarter is actually the midpoint between the Earth's
perihelion and aphelion. This date is approximately October 2nd. The date can be shifted due to trading days or lunar aspects (Full/New Moons, etc) which occur close to this date. Murrey's "birthday, etc" is pure nonsense.

If you have a copy of this particular Gann course you know EXACTLY where "Murrey Math"
came from. I have "Murrey Math" charts which I drew over 10 years before I ever saw the "Murrey Math" manual :-)

Best wishes, Bonnie Lee Hill

PB comment : Nobody has ever publicly seen this specific Gann Course
  • #11,835
  • Dec 11, 2023 6:25pm Dec 11, 2023 6:25pm
a comment from a poster at time about the Starting Date

I think the start date is important. FOM is just giving out some flack since nobody is recognizing the true elements of the discussion. Of course, if the start date is wrong, then it won't be very important, obviously. But, get it right and the results arequite impressive, enough so that it is worth the effort to set it
properly.

First, to counteract some of the FOM sarcasm recently:

I. The Start Date DOES RESET every year in October.

II. The Start DATE IS IMPORTANT.

Check how many significant turns occur within +/-4 trading days of the start of any of the 64 bar frames.

III. It is not related to the first frost or Murrey's birthday


Although both may be close to the date, this is just fluff to conceal the true origin of the date.

I have been testing a proposed start date algorithm for a little while, and I think that it would explain Murrey's start dates for the most part. It may not be perfect, but certainly seems close. It
also seems to agree with most of Murrey's nuggets of wisdom along the way.

Murrey Points to Keep in Mind:

1. It is related to the harvest moon somehow.

2. It is related to multiples of 4 trading days.

3. It should be in the first week of October each year.

4. Once set, the count remains fixed until the next October.

Note, Point #4 was the fooler because some earlier versions of Murrey's EOD program had some bugs related to the date which caused it to slide up or back 1 or 2 days at times, which clouded this issue, even though Murrey would state that the count did not adjust mid-year. But, close examination of Murrey's Trader's World articles shows that this date should not be changed once set until the next October.
  • #11,836
  • Dec 11, 2023 6:30pm Dec 11, 2023 6:30pm
Proposed Start Date Algorithm:

1. Take the first full moon which is on or after Sep 21 as the
HARVEST MOON and denote that date as the anchor for the count.

2. If the full moon occurs earlier than 12:45 ET, then back up 1
calendar day and call that the anchor.

Note, that +8 trading days from this anchor date must be October 1st or later. If the September full moon is not within +8 trading days, you must switch to the October full moon as the anchor.

3. In cases of weekend anchor dates, ALWAYS round BACK not FORWARDS.

The trading frame dates tend to creep larger over the year, since
there are only 252-253 trading days a year, but 256 trading dates in 4 64-bar frames. So, we would always want to round back, not forwards, since the tendency would be to move farther ahead of the actual year, not behind it.

3. If the anchor date is in September, then count either +4 trading days or +8 trading days until reaching a date in October. This date then becomes the year's start date.

4. If the anchor date is in October, then count either 0 trading days, -4 trading days, or -8 trading days until reaching a date which is October 7 or earlier.
  • #11,837
  • Dec 11, 2023 6:32pm Dec 11, 2023 6:32pm
Examples:

1. 2002-2003 Start Date:

Full Moon = Sept. 21 09:59 ET.

Because it is before 12:45 ET, back up 1 calendar day.

Anchor Date = Sept. 20, 2002.

Add 8 trading days = Oct. 2, 2002 ( Murrey Math Start Date ).


2. 2001-2002 Start Date:

Full Moon = Oct. 2 09:49 ET.

Because it is before 12:45 ET, back up 1 calendar day.

Anchor Date = Oct. 1, 2001. ( Murrey Math Start Date ).


3. 2000-2001 Start Date:

Full Moon = Oct. 13 04:53 ET.

Because it is before 12:45 ET, back up 1 calendar day.

Anchor Date = Oct. 12, 2000.

Subtract 4 trading days = Oct. 6, 2000.

** Note, Murrey's start date = Oct. 9, 2000 which is only different by 1 trading day and results from the rounding rule. Backtesting shows that Oct. 6, 2000 shows somewhat better turns, however, so I've stuck with the rounding rules as shown.
  • #11,838
  • Dec 11, 2023 6:36pm Dec 11, 2023 6:36pm
4. 1999-2000 Start Date:

Full Moon = Sept. 25, 1999 06:51 ET.

Because it is before 12:45 ET, back up 1 calendar day.

Anchor Date = Sept. 24, 1999.

Add 8 trading days = Oct. 6, 1999 ( Murrey Math Start Date. )


5. 1998-1999 Start Date:

Full Moon = Oct. 5, 1998 16:12 ET.

No rounding needed.

Anchor Date = Oct. 5, 1998 ( Murrey Math Start Date. )


6. 1997-1998 Start Date:

Full Moon = Oct. 15, 1997 23:46 ET.

No rounding needed.

Anchor Date = Oct. 15, 1997 ( Murrey Math Start Date. )

Subtract 8 trading days = Oct. 3, 1997 ( Murrey Math Start Date. )


7. 1996-1997 Start Date:

Full Moon = Sept. 26, 1996 22:51 ET.

No rounding needed.

Anchor Date = Sept. 26, 1996

Add 4 trading days = Oct. 2, 1996 ( Murrey Math Start Date. )
  • #11,839
  • Dec 11, 2023 6:37pm Dec 11, 2023 6:37pm
Quoting parisboy
a comment from a poster at time about the Starting Date I think the start date is important. FOM is just giving out some flack since nobody is recognizing the true elements of the discussion. Of course, if the start date is wrong, then it won't be very important, obviously. But, get it right and the results arequite impressive, enough so that it is worth the effort to set it properly. First, to counteract some of the FOM sarcasm recently: I. The Start Date DOES RESET every year in October. II. The Start DATE IS IMPORTANT. Check how many significant...
Very Interesting
In the time-warped 2011 version that I can still run, the vertical timinglines are also hard coded into the SW and all timelines are drawn automatically, even when data to 2023 is used.
So 'hard frost' as an explanation from TH is probably a deliberate blurring of info / disinfo
Using the Bonnie Hill info above, for 2023 the dates are Apehelion 6.7.23 & Perhelion 3.1.24, so midpoint = 5.10.23
In the MM SW, this date is the automatically drawn start of the frame, so PBs info seems to be correct
Attached Image (click to enlarge)
Click to Enlarge Name: midpoint.png Size: 52 KB
  • #11,840
  • Dec 11, 2023 6:40pm Dec 11, 2023 6:40pm
8. 1995-1996 Start Date:

Full Moon = Oct. 8, 1995 11:52 ET.

Anchor Date = Oct. 7, 1995 ( Saturday ).

Back to trading day = Oct. 6, 1995 ( Murrey Math Start Date. )


9. 1994-1995 Start Date:

Full Moon = Oct. 19, 1994 08:18 ET.

Anchor Date = Oct. 18, 1994

Subtract 8 trading days = Oct. 6, 1994

** Note, Murrey Math Start Date = Oct. 7, 1994 (rounding difference)
22 traders viewing now