[CANSLIM] Date: Mon, 18 Aug 2003 23:33:23 -0400


From: "Lung Yien The" <ylung@alumni.engin.umich.edu>
Subject: [CANSLIM] Date: Mon, 18 Aug 2003 23:33:23 -0400
Date: Mon, 18 Aug 2003 21:33:33 -0600

Next Article (by Subject): RE: [CANSLIM] Date: Mon, 18 Aug 2003 23:33:23 -0400 "Katherine Malm"
Previous Article (by Subject): [CANSLIM] AUO: What to do " Bob Hodes"
Next in Thread: RE: [CANSLIM] Date: Mon, 18 Aug 2003 23:33:23 -0400 "Katherine Malm"
Articles sorted by: [Date] [Author] [Subject]


This is a multi-part message in MIME format.

- ------=_NextPart_000_0011_01C365E1.1D863ED0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: 7bit

hi guys,
 
I was looking over Breakoutwatch.com and I have some questions about the
formulation of some components of Canslim Evaluator. Would someone help
me clarify some of my questions?
 
1) Relative Strength Rank (RS Rank)
Is this the same rank as employed by IBD? I remember reading from HTMMIS
that relative strength is calculated by taking the 52Week price change
of each stock in the universe and ranked from 1 to 99. Is this the
correct formula or is there a weighting scheme such that recent history
is more important than past history?
 
2) Up/Down Ratio (U/D)
What is the time span in order to generate the Up/Down ratio? 50 days?
10 days? Is there a standard practice?
 
3) Industry Score and Industry Rank Methodology
From what I can understand, the first process is to obtain these numbers
for each stock in the Universe. Then, give a rank from 1 to 99 for each
stock and each item.
 - The current price relative to the 50 day MA
 - The current price relative to the 200 day MA
 - The 50 day MA relative to the 200 day MA
 - The stock's price relative to its 52 Week high
 - the stock's RS rank
 
Next, in order to get the industry score, we take the average of each
item of all the companies that belong to a certain industry or sector -
which gives us a 5 aggregated industry numbers. These numbers are then
aggregated together (maybe taken an average or a weighted average) and
it becomes the industry score.
 
Is my understanding correct?
 
Thanks
Yien 

- ------=_NextPart_000_0011_01C365E1.1D863ED0
Content-Type: text/html;
	charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =
charset=3Dus-ascii">
<TITLE>Message</TITLE>

<META content=3D"MSHTML 6.00.2800.1106" name=3DGENERATOR></HEAD>
<BODY>
<DIV><FONT face=3DArial size=3D2><SPAN class=3D389403902-19082003>hi=20
guys,</SPAN></FONT></DIV>
<DIV><FONT face=3DArial size=3D2><SPAN=20
class=3D389403902-19082003></SPAN></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2><SPAN class=3D389403902-19082003>I was =
looking over=20
Breakoutwatch.com&nbsp;and I have some questions about the formulation =
of some=20
components of Canslim Evaluator. Would someone help me clarify some of =
my=20
questions?</SPAN></FONT></DIV>
<DIV><FONT face=3DArial size=3D2><SPAN=20
class=3D389403902-19082003></SPAN></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2><SPAN class=3D389403902-19082003>1) =
Relative Strength=20
Rank (RS Rank)</SPAN></FONT></DIV>
<DIV><FONT face=3DArial size=3D2><SPAN class=3D389403902-19082003>Is =
this the same=20
rank as employed by IBD? I remember reading from HTMMIS that relative =
strength=20
is calculated by taking the 52Week price change of each stock in the =
universe=20
and ranked from 1 to 99. Is this the correct formula or is there a =
weighting=20
scheme such that recent history is more important than past=20
history?</SPAN></FONT></DIV>
<DIV><FONT face=3DArial size=3D2><SPAN=20
class=3D389403902-19082003></SPAN></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2><SPAN class=3D389403902-19082003>2) =
Up/Down Ratio=20
(U/D)</SPAN></FONT></DIV>
<DIV><FONT face=3DArial size=3D2><SPAN class=3D389403902-19082003>What =
is the time=20
span in order to generate the Up/Down ratio? 50 days? 10 days? Is =
there&nbsp;a=20
standard practice?</SPAN></FONT></DIV>
<DIV><FONT face=3DArial size=3D2><SPAN=20
class=3D389403902-19082003></SPAN></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2><SPAN class=3D389403902-19082003>3) =
Industry Score=20
and Industry Rank Methodology</SPAN></FONT></DIV>
<DIV><FONT face=3DArial size=3D2><SPAN class=3D389403902-19082003>From =
what I can=20
understand, the first process is to obtain these numbers for each stock =
in the=20
Universe. Then, give a rank from 1 to 99 for each stock and each=20
item.</SPAN></FONT></DIV>
<DIV><FONT face=3DArial size=3D2><SPAN =
class=3D389403902-19082003>&nbsp;- The current=20
price relative to the 50 day MA</SPAN></FONT></DIV>
<DIV><FONT face=3DArial size=3D2><SPAN =
class=3D389403902-19082003>&nbsp;- The current=20
price relative to the 200 day MA</SPAN></FONT></DIV>
<DIV><FONT face=3DArial size=3D2><SPAN =
class=3D389403902-19082003>&nbsp;- The 50 day=20
MA relative to the 200 day MA</SPAN></FONT></DIV>
<DIV><FONT face=3DArial size=3D2><SPAN =
class=3D389403902-19082003>&nbsp;- The stock's=20
price relative to its 52 Week high</SPAN></FONT></DIV>
<DIV><FONT face=3DArial size=3D2><SPAN =
class=3D389403902-19082003>&nbsp;- the stock's=20
RS rank</SPAN></FONT></DIV>
<DIV><FONT face=3DArial size=3D2><SPAN=20
class=3D389403902-19082003></SPAN></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2><SPAN class=3D389403902-19082003>Next, =
in order to=20
get the industry score, we take the average of each item of all the =
companies=20
that belong to a certain industry or sector - which gives us a 5 =
aggregated=20
industry numbers. These numbers are then aggregated together (maybe =
taken an=20
average or a weighted average) and it becomes the industry=20
score.</SPAN></FONT></DIV>
<DIV><FONT face=3DArial size=3D2><SPAN=20
class=3D389403902-19082003></SPAN></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2><SPAN class=3D389403902-19082003>Is my =
understanding=20
correct?</SPAN></FONT></DIV>
<DIV><FONT face=3DArial size=3D2><SPAN=20
class=3D389403902-19082003></SPAN></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2><SPAN=20
class=3D389403902-19082003>Thanks</SPAN></FONT></DIV>
<DIV><FONT face=3DArial size=3D2><SPAN=20
class=3D389403902-19082003>Yien&nbsp;</SPAN></FONT></DIV></BODY></HTML>

- ------=_NextPart_000_0011_01C365E1.1D863ED0--


- -
- -To subscribe/unsubscribe, email "majordomo@xmission.com"
- -In the email body, write "subscribe canslim" or
- -"unsubscribe canslim".  Do not use quotes in your email.

Next Article (by Subject): RE: [CANSLIM] Date: Mon, 18 Aug 2003 23:33:23 -0400 "Katherine Malm"
Previous Article (by Subject): [CANSLIM] AUO: What to do " Bob Hodes"
Next in Thread: RE: [CANSLIM] Date: Mon, 18 Aug 2003 23:33:23 -0400 "Katherine Malm"
Articles sorted by: [Date] [Author] [Subject]


Go to Jeff Salisbury LWGate Home Page.