Rank: Advanced Member
Groups: Registered, Registered Users
Joined: 3/19/2005(UTC)
Posts: 2,995
Was thanked: 14 time(s) in 10 post(s)
|
Still working with the Greg Morris indicators... So far I'm stuck on the McClellan Oscillator which is basically step 1 for me. I am having trouble getting a consistent value between indicators and/or data. Let's start with the basic MetaStock indicator:
BM NASDAQ McClellan Oscillator
[code:1:61e86a17d7]{This indicator calculates the McClellan Oscillator using data from the NASDAQ exchange.}
AmD:=Security("C:\\MetaStock Data\\BM Data\\us&advq",C);
Mov(AmD,19,E)-Mov(AmD,39,E)[/code:1:61e86a17d7]
Note that it is referencing a QC symbol which is current as of 11/21/05. The value for 11/21/05 = 57.75. Plot the same indicator using the Reuters Datalink symbol (x.nasd-a) and the value changes to 58.78 for the same date. Well, the values are fairly close and since the formula is exactly the same, the value difference has to be in the data. Hmm...
Now, let's look at Greg Morris' indicator:
REU Nasdaq Daily McClellan Oscillator - 19-39
Code is password protected, but it uses the Reuters symbol (x.nasd-a). I know this because of the REU prefix in the title of the indicator AND because if it isn't present in the BM Data folder, the indicator will give you an error message. Fair enough, but if I plot it using the same data as above, it returns a value of 114.85 for 11/21/05. Wow! What a difference. I thought the whole purpose of using the EMA was to allow similar values to be duplicated...
The difference in the 2 plots is primarily a difference in scale. If you overlay one on top of the other with the second indicator having no scale, the patterns are 99% identical, but not exactly; see the attached graph. So, I started looking into what the value of the oscillator should be to see which indicator was correct and all I could find was the daily values listed at http://stockcharts.com/charts/indices/McSumNASD.html
|
|
|
|
|
|
Rank: Newbie
Groups: Registered, Registered Users
Joined: 6/1/2005(UTC)
Posts: 8
|
George,
I included a handful of McClellan Indicators in the package for the very reason that you are asking questions. There are a number of differenet methods of calculation, in fact, Tom and Sherman McClellan use two different ones. Some are the 19 and 39 difference of exponential averages of the net advances (A-D), some use a ratio of (A-D)/(A+D), some use (A-D)/Total Issues. Plus, when you get into the McClellan Summation Index, the +1000 originally used by the McClellans, is also changed to be zero. And I didn't even mention the Miekka formula adjustment. I just had an article published in Stocks & Commodities that addresses a lot of this. Hope this helps explain why you might see some differences, you need to ensure you are using the same concept and formula.
Greg Morris
|
|
|
|
|
|
Rank: Newbie
Groups: Registered, Registered Users
Joined: 6/1/2005(UTC)
Posts: 8
|
George,
Also, different data services use different breadth data. Reuter's breadth data does not include warrants, rights, and some other issues, where other services, such as Dial Data and Pinnacle include all issues. If I had my way, we would only use breadth data of common stocks only. Greg
|
|
|
|
|
|
Rank: Advanced Member
Groups: Registered, Registered Users
Joined: 3/19/2005(UTC)
Posts: 2,995
Was thanked: 14 time(s) in 10 post(s)
|
Thanks Greg for the personal reply. That just made my day! -)
I knew my confusion grew from a lack of understanding, but I couldn't seem to find the answer. I just got my TASC today after returning home from vacation. I'll surely check out your work there too.
In your book, on page 77 you recount Sherman's range of -125 to -130 as a bearish indicator. Which version of the formula does this range apply to? With your formula "REU Nasdaq Daily McClellan Oscillator - 19-39" that seems like an appropriate level from a glance, but using the Equis formula "AmD:=Security("C:\\MetaStock Data\\BM Data\\us&advq",C); Mov(AmD,19,E)-Mov(AmD,39,E)" that range becomes useless. I guess what I'm looking for is some idea as to a reasonable range of values for each of the indicators in your toolbox.
Along the same lines, the summation has the same problem. Your indicator "REU Nasdaq Daily McClellan Summation" plots +823.32 on the chart above and the Equis formula "AmD:=Security("C:\\MetaStock Data\\BM Data\\us&AMDQ",C); Cum(Mov(AmD,19,E)-Mov(AmD,39,E) )" plots -3566. Two completely different indicators.
|
|
|
|
|
|
Rank: Advanced Member
Groups: Registered, Registered Users, Subscribers
Joined: 10/29/2004(UTC)
Posts: 1,394
Location: Glastonbury, CT
Was thanked: 2 time(s) in 2 post(s)
|
George,
Equis is not going to like this but reuters datalink gives different values than Quotecenter
Compare the nasdaq new highs using Reuters and then the nasdaq new highs using Quotecenter. You will get two different values.
I would think that the markets would give out the same value for the current day. Which one is correct?
|
|
|
|
|
|
Rank: Advanced Member
Groups: Registered, Registered Users
Joined: 3/19/2005(UTC)
Posts: 2,995
Was thanked: 14 time(s) in 10 post(s)
|
Quote:reuters datalink gives different values than Quotecenter
Exactly.
Thanks Henry, I know that you've done a lot of evaluations in this area; previously, you helped me with advances and declines and made note of it. I guess it didn't sink in until I heard it for a second time.
I don't want to get too bogged down with exact values, because the whole purpose is a larger picture. Still, it is a little annoying to us OCD types.
|
|
|
|
|
|
Rank: Advanced Member
Groups: Registered, Registered Users, Subscribers
Joined: 9/9/2004(UTC)
Posts: 107
Location: Salt Lake City, UT
|
We are checking it out. I will let you know what I find.
Doc
|
|
|
|
|
|
Rank: Newbie
Groups: Registered, Registered Users
Joined: 6/1/2005(UTC)
Posts: 8
|
All of the information in my book was in reference to the NYSE based indicators, not the Nasdaq. Hope that helps.
Also, Reuters does not include all issues for their breadth data. This is the reason I insisted that Pinnacle be included in the package as they do include all issues for their breadth numbers. I do not think picking and choosing what to include or not include is a good thing. I would love anyone who would make breadth data available for common stocks only. AND provide a large amount of historical data also. Greg
PS I have to add that I have no idea which is better, only that starting with the purist form is usually the best method.
|
|
|
|
|
|
Rank: Advanced Member
Groups: Registered, Registered Users
Joined: 3/19/2005(UTC)
Posts: 2,995
Was thanked: 14 time(s) in 10 post(s)
|
Thanks Greg.
Have a great holiday!
|
|
|
|
|
|
| Users browsing this topic |
|
Guest (Hidden)
|
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.
