|
PureBytes Links
Trading Reference Links
|
Hi Pablo
I think the mistake is your's, though not one of the possibilities I suggested. This code is counting events (divergences), so it follows that to get the same result from the Explorer as the Indicator Builder you MUST compare apples with apples (use the same number of bars). The Explorer only reports to 4 decimal digits so this value may be rounded (and therefore different) from the value on a chart via the data window. This however is an issue that most of us can live with, and it's not related to your problem.
If MetaStock is set to open 2500 bars for a chart then you must explore 2500 bars also to be sure of capturing the same number of divergences and generating the same results from the exploration. Even a difference of one bar could legitimately give a different result. This is a fact of life when accumulating data - your starting point is (date, time or bar) has a big influence on the final result. The earlier you start accumulating the greater the probability that you will accumulate more divergences.
This comes back to what Preston told you - "Make sure that the displayed data and the explored data are the same". In many situations it's reasonable to expect that an accurate result is possible by exploring many fewer bars than usually required for an open chart, but the particular formula you're using demands that you use exactly the same number of bars as the chart you're are comparing the result with. To do otherwise, as you have proved, generates different values from the accumulations and therefore different results.
I don't need to be a defender of MetaStock. MS is not perfect, but most reported problems from new users are not the result of MetaStock making the mistakes.
Regards
Roy
www.metastocktips.co.nz
----- Original Message -----
From: Paolo
To: Roy Larsen
Sent: Monday, May 29, 2006 1:37 PM
Subject: [Norton AntiSpam] [SPAM] [EquisMetaStock Group] "what am I doing wrong?"
Roy,
THANK U VERY MUCH for your suggestions. They are appreciated!
I've created my explorer "di novo", using default peridicity, records
to be loaded, etc. I've checked ALL.
It's true I DO NOT TRUST in Metastock because it's new for me.
So.... MAYBE I should ask myself "what am I doing wrong?" as you said.
Anyway, I guess there is something wrong with Metastock.....
If you are a *Metastock defender*, I invite you to test my code.
If you are able to pick the same values from the "indicator" and the
"explorer" then I'll congratulate you :)
... and I'll become very happy man (no more conflicts!)
I still thinking ANY CODE should make the same results INDEPENDENTLY
if it's in the Indicator Builder or Explorer (!)
The code to try is:
ColA:
====
{ User inputs }
pds:=25; {for calculus of divergences}
p1:=20; { Divergence pds for short moving average }
p2:=200; { Divergence pds for long moving average }
s1:= Mov(C,p1,E);
s2:= Mov(C,p2,E);
{ Divergences }
divs:=If(s1<Mov(s1,pds,E) AND (s1<Ref(s1,-pds)) AND
s2>Mov(s2,pds,E) AND (s2>Ref(s2,-pds)),-1,
If(s1>Mov(s1,pds,E) AND (s1>Ref(s1,-pds)) AND
s2<Mov(s2,pds,E) AND (s2<Ref(s2,-pds)) ,1,0));
DP:=If(divs=1,divs+PREV,PREV); { acumulate of (+)divs }
DN:=If(divs=-1,divs+PREV,PREV); { acumulate of (-)divs }
Abs(DP/DN); {Ratio}
Filter: (of course, NOT in the indicator)
=======
ColA<>0
--------
Thank U in advance!
Best reagards,
Pablo
RL> Hi Paolo
RL> I suggest that you Edit your exploration, look under Options, and set Exploration Date to Most Recent Date. Each exploration has this choice, and if you created your exploration by copying from
RL> another that was set to Specific Date then the new exploration will have the same setting.
RL> Another distinct possibility is that you have Periodicity set incorrectly - this will also give values that are way off.
RL> The one thing that you can be sure of is that your exploration is doing exactly what you have told it to do. You are responsible for errors in the results, and accepting that will enable you to
RL> follow through and fix mistakes. Assuming that the problem lies with the Explorer makes finding the problem much harder because you'll be looking in the wrong places. I'm not suggesting that you
RL> personally are blaming MetaStock, but offering my thoughts for those that whose first thought is such a situation is to ask "what is MetaStock doing wrong?" when the question should be "what am I
RL> doing wrong?".
RL> Regards
RL> Roy
RL> www.metastocktips.co.nz
RL> ----- Original Message -----
RL> From: Paolo
RL> To: pumrysh
RL> Sent: Monday, May 29, 2006 5:14 AM
RL> Subject: [Norton AntiSpam] [SPAM] Re: [EquisMetaStock Group] Re: Explorer (!)(!)(
RL> Preston:
RL> You was RIGHT about explorer wasn't loading ALL the records
RL> BUT now after load 10,000 records, with other values, the PROBLEM
RL> STILL (continue)
RL> The values are so diferent as: 1.5061 (explorer) vs. 0.083(indicator)
RL> I thought after load ALL the records the problem wold be solved but
RL> now I'm afraid it has no solution.
RL> It's crazy the explorer takes values from 1996 when the stock data of
RL> that security (CEPU2) starts in 2004 !!!!!!! :O
RL> Thank U for help me.
RL> Best regads,
RL> Pablo
p>> You explored a different length of data. Make sure that the
p>> displayed data and explored data are the same. When you first open
p>> the explorer window, you will notice an explore, reports, and
p>> options button at the bottom. Go into options to set the looback
p>> period.
p>> Preston
p>> --- In equismetastock@xxxxxxxxxxxxxxx, Paolo <italoarg76@xxx> wrote:
>>>
>>> Hi!!!
>>> I've developed an indicator and then I put it into
>>> the explorer (as Jose Silve teach me)
>>> The values calculated by the explorer are VERY diferent of the
>>> indicator and I could to see the explorer values were WRONG!!!
>>> What's happen with the explorer ?!!
>>> How can I use this tool if it fails so much ?!!!!!!!!!!!!!!!!!
>>> The explorer code tried was:
>>>
>>> ColA:
>>> ====
>>>
>>> { User inputs }
>>> pds:=25; {for calculus of divergences}
>>> p1:=20; { Divergence pds for short moving average }
>>> p2:=200; { Divergence pds for long moving average }
>>>
>>> s1:= Mov(C,p1,E);
>>> s2:= Mov(C,p2,E);
>>>
>>> { Divergences }
>>>
>>> divs:=If(s1<Mov(s1,pds,E) AND (s1<Ref(s1,-pds)) AND
>>> s2>Mov(s2,pds,E) AND (s2>Ref(s2,-pds)),-1,
>>> If(s1>Mov(s1,pds,E) AND (s1>Ref(s1,-pds)) AND
>>> s2<Mov(s2,pds,E) AND (s2<Ref(s2,-pds)) ,1,0));
>>>
>>> DP:=If(divs=1,divs+PREV,PREV); { acumulate of (+)divs }
>>> DN:=If(divs=-1,divs+PREV,PREV); { acumulate of (-)divs }
>>>
>>> Abs(DP/DN); {Ratio}
>>>
>>> Filter:
>>> ======
>>>
>>> colA<>0
>>>
>>> -------
>>> Thank U in advance.
>>> Arrivederci!
>>> Pablo
>>>
p>> Yahoo! Groups Links
RL> Pablo Bozzolo
RL> ---
RL> Contact info:
RL> Movil: 221 5384617
RL> E-mail: italoarg76@xxxxxxxx
RL> MSN: italoarg76@xxxxxxxxxxx
RL> ICQ: 77207033
RL> SPONSORED LINKS Business finance course Business to business finance Small business finance
RL> Business finance consultant Business finance magazine Business finance schools
RL> --------------------------------------------------------------------------------
RL> YAHOO! GROUPS LINKS
RL> a.. Visit your group "equismetastock" on the web.
RL> b.. To unsubscribe from this group, send an email to:
RL> equismetastock-unsubscribe@xxxxxxxxxxxxxxx
RL> c.. Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.
RL> --------------------------------------------------------------------------------
RL> [Non-text portions of this message have been removed]
RL> Yahoo! Groups Links
Pablo Bozzolo
---
Contact info:
Movil: 221 5384617
E-mail: italoarg76@xxxxxxxx
MSN: italoarg76@xxxxxxxxxxx
ICQ: 77207033
SPONSORED LINKS Business finance course Business to business finance Small business finance
Business finance consultant Business finance magazine Business finance schools
--------------------------------------------------------------------------------
YAHOO! GROUPS LINKS
a.. Visit your group "equismetastock" on the web.
b.. To unsubscribe from this group, send an email to:
equismetastock-unsubscribe@xxxxxxxxxxxxxxx
c.. Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.
--------------------------------------------------------------------------------
[Non-text portions of this message have been removed]
------------------------ Yahoo! Groups Sponsor --------------------~-->
Get to your groups with one click. Know instantly when new email arrives
http://us.click.yahoo.com/.7bhrC/MGxNAA/yQLSAA/BefplB/TM
--------------------------------------------------------------------~->
Yahoo! Groups Links
<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/equismetastock/
<*> To unsubscribe from this group, send an email to:
equismetastock-unsubscribe@xxxxxxxxxxxxxxx
<*> Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/
|