From: "Katherine Malm" <kmalm@earthlink.net>
Subject: Re: [CANSLIM] How to compute the relative strength?
Date: Thu, 19 Sep 2002 20:27:29 -0500
Previous Article (by Date): RE: [CANSLIM] How to compute the relative strength? "Mike Gibbons"
Top of Thread: [CANSLIM] How to compute the relative strength? "Quan Tran"
Articles sorted by: [Date]
[Author]
[Subject]
Hi Quan,
QP2 uses the same methodology, but gets ever so slightly different rankings
because their total universe of stocks is not exactly the same as IBD's.
It's close enough for mining for CANSLIM stocks, so I don't think there's
any compelling reason not to use QP2's numbers.
Here's the skinny on the RS Rank:
RS is a gauge of a stock's price performance vs. all other stocks in the IBD
database (approx. 10,000 stocks). The change in each stock's price for each
of the previous four quarters is used to derive a weighted price performance
for each stock. This weighted price performance is then used to rank all
stocks against each other. Each stock is then given an RS Rank from 99
(highest) to 1 (lowest). This value is updated daily, so is a dynamic
indicator of relative performance against all other stocks given the latest
closing prices.
Formula for determining RS:
Weighted Price Change = (Price change this quarter * 40%)
+ (Price change 1 qtr ago * 20%)
+ (Price change 2 qtrs ago * 20%)
+ (Price change 3 qtrs ago * 20%)
If there are less than 4 quarters of data, then the weighting for 3 quarters
is (50/30/20) and for 2 quarters it's (60/40). I'm not sure how IBD handles
situations where there is less than 1 quarter of data, but I know that QP2
just doesn't give them a rank until there's at least 3 months price data.
There are also some who believe that IBD "plays" with their RS Rankings a
bit and uses some of the values as "holding bins" but I'm disinclined to
believe that, as there is no value in futzing with the numbers. RS Rank
simply gives you a way of separating laggards and leaders over a year's
time. In the near term, the RS Line (which simply measure strength relative
to the market as a whole) is far more useful for monitoring
stock price and relative strength behavior.
- --Katherine
- ----- Original Message -----
From: "Quan Tran" <quantran@sbcglobal.net>
To: <canslim@lists.xmission.com>
Sent: Thursday, September 19, 2002 7:05 PM
Subject: [CANSLIM] How to compute the relative strength?
| Hi all,
|
| Sorry if this question has been asked before. I am using QP, and
| I know that it computes RS differently than IBD. Does anyone know
| how IBD compute RS? The reason for this question is that I'd like
| to use IBD calculation as a filter in my QP scan.
|
| -Quan
| quan@sbcglobal.net
|
| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| Income tax has made more liars out of the American people
| than golf.
|
| - Will Rogers
| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
|
| -
| -To subscribe/unsubscribe, email "majordomo@xmission.com"
| -In the email body, write "subscribe canslim" or
| -"unsubscribe canslim". Do not use quotes in your email.
- -
- -To subscribe/unsubscribe, email "majordomo@xmission.com"
- -In the email body, write "subscribe canslim" or
- -"unsubscribe canslim". Do not use quotes in your email.
Previous Article (by Date): RE: [CANSLIM] How to compute the relative strength? "Mike Gibbons"
Top of Thread: [CANSLIM] How to compute the relative strength? "Quan Tran"
Articles sorted by: [Date]
[Author]
[Subject]