|
PureBytes Links
Trading Reference Links
|
Sorry I didn't get back on this earlier. The possible problem of keeping the
two steps separate is that they are not independent variables. i.e. an equal
dollar relative strength (calculated as a spread) will give different
results (and different rankings) than one based on volatility - simply
because the spread ratios are different. That is, on an equal dollar basis
you will have a larger quantity of the more volatile security. If you look
at the relative strength of the NDX vs. SPX, it will tell you to always be
long or short the NDX, never the SPX, because the NDX is more volatile (and
has a higher beta, if that's the way you look at it). A lot of relative
strength work just puts you into higher beta or more volatile securities
when the market goes up (if you are long) - which is fine, as long as the
user understands that they are increasing portfolio risk as a result.
If those high RS/high vol candidates no longer look good on a vol adjusted
basis, then they may not belong in the portfolio if you are trying to keep a
constant vol.
What isn't considered here is correlation and covariance within the
portfolio (i.e. that high vol stock may actually reduce the volatility of
the portfolio if its negatively correlated) - I believe that Morningstar has
some good portfolio risk tools, but I haven't tried them.
Andrew
-----Original Message-----
From: equismetastock@xxxxxxxxxxxxxxx [mailto:equismetastock@xxxxxxxxxxxxxxx]
On Behalf Of metastkuser
Sent: Tuesday, June 07, 2005 7:05 PM
To: equismetastock@xxxxxxxxxxxxxxx
Subject: [EquisMetaStock Group] Re: Volatility adjusted relative strength
Andrew,
Interesting thought. I'm doing quite a bit right now with volatility
-- for setup, entry, exit and money management -- and had not thought of
this.
The result of normalizing the RS using beta or atr would suggest a "risk
adjusted" metric for comparing one security's RS against another's.
I'm not sure if that is worthwhile. I like to know the "raw" RS first. Then
I do my "risk adjustments" (i.e. volatility-based position sizing and money
management stop placement) as a next step. Merging these two steps might
fail to highlight some high RS/high volatiltiy (but
promising) candidates?
--- In equismetastock@xxxxxxxxxxxxxxx, "Andrew Tomlinson"
<andrew_tomlinson@xxxx> wrote:
>
> I've been playing around with calculating relative strength not as
C/index
> but as a volatility adjusted spread. This more nearly approximates the
> relative performance assuming an equal amount of risk-adjusted capital
> invested in each security. So, for example, rather than a regular
relative
> strength vs. the S&P500
>
> SP:=Security("C:\MetaStock Data\Indeces\.spx",C); RS:=C/SP;rs;
>
> I would take the spread between the two securities adjusted for their
> volatility expressed as ATR as % of price e.g.
>
> P1:=50;
> S1H:=H;
> S1L:=L;
> S1C:=C;
> S2H:=Security("C:\MetaStock Data\Indeces\.spx",H);
> S2L:=Security("C:\MetaStock Data\Indeces\.spx",L);
> S2C:=Security("C:\MetaStock Data\Indeces\.spx",C);
> TR1:=Max(Max(S1H-S1L,S1H-Ref(S1C,-1)),(Ref(S1C,-1)-S1L));
> ATR1:=Mov(TR1,P1,S)/Mov(S1C,P1,S);
> TR2:=Max(Max(S2H-S2L,S2H-Ref(S2C,-1)),(Ref(S2C,-1)-S2L));
> ATR2:=Mov(TR2,P1,S)/Mov(S2c,P1,S);
> Ratio:=ATR1/ATR2;
> Spread:=S1C-S2C*ratio;spread;
>
> I'm curious - has anyone else looked at this? Theoretically, this
would be
> an alternative to any regular relative strength calculation, and would
> better reflect the reality of "am I better off investing in A or
B?". I have
> also seen this done using beta as the volatility measure.
>
> Andrew
Yahoo! Groups Links
------------------------ Yahoo! Groups Sponsor --------------------~-->
Has someone you know been affected by illness or disease?
Network for Good is THE place to support health awareness efforts!
http://us.click.yahoo.com/Rcy2bD/UOnJAA/cosFAA/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/
|