How to standardize RS?

B

Babak

I know this may seem like a really dumb question but here goes...

I'm trying to 'standardize' relative strength (NOT the RSI but RS) so that it bobs and weaves along an axis.

RS is security price divided by index level, now what do you have to do to it to get the standardized version?

I know there is a way to do this since I've seen charts where RS is standardized but I've yet to run across how they do this (maybe because it is so simple!).

For example, if you have Pring's Market Momentum, look on page 311. Weinstein's book also has charts with standardized RS all over it.

Thanks guys. :(
 
Quote from Babak:

I.....
RS is security price divided by index level, now what do you have to do to it to get the standardized version?
.....
More...

I don't have those books, so an example would be helpful.

On the other hand: price divided by index level is the standardized version because what you are doing is calculating the security price as percentage of the index level. This allows you to compare the RS of different securities.

Or do you mean the RS relative to a certain date? So the RS starts as 1 (=100%) on a certain date and the chart shows the relative development since that date?

That's: {(security price now)/(security price then)}/{(index price now)/(index price then)}

Hth

Bernd Kuerbs
 
If RS = Price / Index, then (as per Bernd's suggestion),
you can see whether it's going up or down by using:

RS(now) / RS(then)

which might range from 0.95 to 1.1 (for example).

To get the percentage change, you could use:
100 [ RS(now) / RS(then) - 1 ]

which would oscillate between -5% to +10% (for example).
 
Thanks guys, yes, I think we're on the right track. But even with the comparison to 'then' it doesn't produce an oscillation around an axis.

I think it might be necessary on top of that to compare to a changing 'then'. Kind of a ROC RS, if you get my drift. IOW you're comparing today's RS to say the RS you had 20, 50, etc. days ago.

As for providing an example, I don't know how to do that (other than scanning the page of the books I mentioned). I've searched high and low on the net but haven't found anything re this. Maybe its not called 'standardized' RS as I've named it in this thread.
 
Babak:
I think you'll need to define "RS" before you get an acceptable answer.

The "standard" definition(s) of RS will not guarantee oscillations about an axis ... except by accident :D

(It's sorta like asking how to make S&P500 returns oscillate about an axis.)
 
Quote from Babak:

....(other than scanning the page of the books I mentioned). I've searched high and low on the net but haven't found anything re this. ...
More...

Your description only confuses me.

The chart shows SPX, MSFT since 5th January as well as the relative strength of SPX and MSFT since that date and MSFT relative to SPX.

Does your chart look like this?

Regards

Bernd Kuerbs
 

Attachments

Quote from Babak:

Thanks guys, yes, I think we're on the right track. But even with the comparison to 'then' it doesn't produce an oscillation around an axis.

I think it might be necessary on top of that to compare to a changing 'then'. Kind of a ROC RS, if you get my drift. IOW you're comparing today's RS to say the RS you had 20, 50, etc. days ago.

As for providing an example, I don't know how to do that (other than scanning the page of the books I mentioned). I've searched high and low on the net but haven't found anything re this. Maybe its not called 'standardized' RS as I've named it in this thread.
More...

I did a lot with that a while ago. Go stock/index, create an ma, say 5 period. Then subtract stock/index from the smooth values and it will oscillate around zero.


John
 
ok I had a bit of a brainstorm. I looked through Weinstein's book and found the company that did those charts and it turns out they are still in the biz:

http://www.mansfield-charts.com/Targets/Samples/FWeeklySamp.htm

Look at #12 on that chart. Its the RS but its not just price/index as you'll notice. You'll have to scroll at the bottom window or just click on #12.

More examples:

http://www.mansfield-charts.com/Targets/Samples/FOTC23Samp.htm
http://www.mansfield-charts.com/Targets/Samples/FMonthlySamp.htm
(click on #10 for these)

http://www.mansfield-charts.com/Targets/Samples/FIntSamp.htm
(click on #20 for this)


They seem to have a formula:

mkt price/ (base price * SP500)

what does that mean?

:confused:
 
Some interesting mathematical machinations are required to make
(MktPrice) / (BasePrice * S&P500)
oscillate about 0:

(MktPrice) / (BasePrice * S&P500) = 0
when ... what?

MktPrice = 0 ?
BasePrice = infinity ?
S&P500 = infinity ?

That must be the "new math" :)
 
I've sent the following e-mail to mansfield (who, I assume, generates the charts):
--------------------------------

I'm confused.

Here:
http://www.mansfield-charts.com/Targets/Samples/FWeeklySamp.htm
you have a chart (#12) which oscillates about the value 0.

It's (apparently) calculated via:
(MktPrice)/(BasePrice * S&P500)

So it's 0 when ... what?

MktPrice = 0 ?
BasePrice = infinity ?

Cheers

--------------------------------
I suspect that the chart actually plots something akin to:
MktPrice / S&P500 - BaseRatio
where BaseRatio is the now-infamous
RS(then)
 
You must log in or register to reply here.
Back
Top