|
PureBytes Links
Trading Reference Links
|
In my experience, fib retracements usually, but not always, are
better log. For Andrews channels the way I use them, linear is
usually better, but I try both and pick the best for the market. For
instance, I maintain monthly fx charts with Andrews-type analysis;
most are linear, but some are log and for a couple there is
sufficient value that I maintain both. I don't use Gann/speed
resistance angles, but when I used to, I think the answer would have
been the same as for Andrews. Some things are no brainer log --
high volatility stocks -- fib and andrews work beautifully log. I
would have thought the same for volatile commodity markets like NG,
but to me the evidence is less clear there.
Another thing on this topic -- log indicators. if you use indicators
like momentum or macd, you might want to try log versions where log
scale makes sense -- especially if you watch divergences. Log scale
macd for instance will reveal divergences in an up market that
regular will not see. Going down, the opposite is true, linear will
show divergences that are not there on log. If anyone is interested,
I'd be happy to post my log Macd code.
Chris
--- In realtraders@xxxx, "wavemechanic" <wd78@xxxx> wrote:
> Stig:
>
> The rule of thumb that I have seen in various books is that
arithmetic is OK for periods less than one year (difference between
the two is small), but that semi-logarithmic should be used above
that (or just use semi-log all the time). I don't know why one would
use semi-log for extended price ranges unless the time factor comes
into play. As for which trendline is right, I don't know of any
study that has established which gives the best signals, and that
should be the determining factor. Perhaps someone has more
information.
>
> Bill
> ----- Original Message -----
> From: Stig O
> To: realtraders@xxxx
> Sent: Sunday, May 20, 2001 6:41 PM
> Subject: Re: [RT] The Gold Thread
>
>
>
>
> Stig:
>
> The picture is a bit different if semi-logarithmic scale is
used, which is appropriate for extended times.
>
> Bill
>
> Interesting point Bill, I have to think about that one.
>
> Do you mean to say that one is right and one is wrong?
>
> Or could one say: "Hey, look at this one, it spells danger
(arithmetc scale). Don't trust you semilog scale.
>
> Or should one say"I use semilog scale, the buy/signal that
shows up on you arithmetic scale has no significance.
>
> What I have learned is to use semi log scale for extended
*price ranges*. I haven't heard the argument with regard to time
before.
> Interesting...
>
>
> ( I tried to send this as private mail, but my browser wouldn't
accept your adress)
>
> Best regards
> Stig
>
> Other inputs on the subject are welcomed.
>
> Just got this link from someone else by the way. Something for
our stocktrader members.
> http://www.investertech.com/tkchart/tkchart.asp?
stkname=drooy+hgmcy+gold+nem+hm+asa+pdg+%24xau.x&px=2&wt=0
>
> Yahoo! Groups Sponsor
>
>
>
>
>
>
> To unsubscribe from this group, send an email to:
> realtraders-unsubscribe@xxxx
>
>
>
> Your use of Yahoo! Groups is subject to the Yahoo! Terms of
Service.
To unsubscribe from this group, send an email to:
realtraders-unsubscribe@xxxxxxxxxxxxxxx
Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
|