|
PureBytes Links
Trading Reference Links
|
<x-html><!DOCTYPE HTML PUBLIC "-//W3C//DTD W3 HTML//EN">
<HTML>
<HEAD>
<META content=text/html;charset=iso-8859-1 http-equiv=Content-Type>
<META content='"MSHTML 4.72.3509.100"' name=GENERATOR>
</HEAD>
<BODY bgColor=#ffffff>
<DIV>
<P> I wanted to exchange some ideas on setting targets
and stops based on short term trend channels. First let me say that I always set
my stops this way. However, I often set my targets based on old highs or lows,
but that is fairly obvious and not the subject of this discussion. Also, for
this discussion I’ll talk about up trend channels and long positions, but
down trend channels and short positions are just the mirror image.</P>
<P> I’ve been experimenting with using the
MetaStock Standard Deviation Channel for setting targets and stops. The main
problem with most trend channels is deciding where to start and end them. I
don’t like the idea of just moving the end points around until the channel
looks right, that’s way to subjective. I want any system I use to be very
objective and not subject to interpretation. After some experimenting, I came up
with the following methodology. I start immediately to the left of the lowest
low in the data being considered and end immediately to the right of the highest
high in the time frame. Then I set the deviation at 2 and extend the channel to
the right. Anytime a new high is hit, I'll drag the right end of the channel
just to the right of that new high. Using this methodology, I know that the
trend channels will be constructed the same way every time. </P>
<P> Now for setting the targets, I’ve been
experimenting with setting the deviation at 2 and setting the target at the
whole number just under the top of the channel. I like to use the whole numbers
because specialists like to take a stock to whole numbers to shake out the stops
and limits. I then move it up to the next whole number just as soon as the
channel will let me. I set the deviation at 2 because I like to let my profits
run and not close them because they hit an arbitrary target. A channel with the
deviation set at 2 will contain 95% of the stock moves. Therefore, the odds are
that if a stock moves outside the top of the channel it will correct soon. Of
course I’ll miss an occasional rocket, but I’m willing to pay that
price to have most of my targets near the stock top.</P>
<P> For setting the stops, I set the deviation at 1 and
set my stop at the x ¾ point just under the bottom of the channel. The
reason for the ¾ number under the channel is again that specialist like
to drive a stock to whole numbers, so a stop ¼ under a whole number is
less likely to be hit. The reason for setting the deviation at 1 is that will
contain 67% of the stock moves. I want my stop to trigger quicker in the bottom
half of the channel than my target does in the to for two reasons. First, the
long-term market bias is up and I reinforce that by only taking long positions
in stocks that are trending up when the market is trending up. Therefore it
makes sense to let a stock run further on the upside than on the downside under
these conditions. Second and more import, the stop is there for protection
– to protect a profit or protect against a large loss. I want that
protection to kick in as soon as possible while still minimizing whip saws.
Picking a point that will contain 67% of the stock moves seems like a reasonable
compromise.</P>
<P> Another point I want to make is the amount of data
necessary for constructing the STUTC. I think the mistake I made on MSFT this
week was that I changed my channel starting point from the 10/08/98 low to the
11/16/98 low which I thought was the pivot point for the latest breakout. The
problem with this was that there was less than two weeks of data without and
significant reaction low. Therefore MSFT closed below my mental stop Thursday.
Luckily, Friday’s action kept me in MSFT. If I hadn’t moved my
starting point, MSFT would have remained well above the stop from that channel.
I’m going to make the general rule that I need at least 21 days of data to
construct a STUTC. I can fudge on that a little if there has been a good
reaction low in the data. I can ignore the rule entirely if the stop is on a
stock for which I haven’t set a target and I want to tighten the stop to
protect a large profit. If I do that, I should recognize that the cost could be
a false stop and I should be willing to re-enter the stock when it sets a new
high. </P>
<P> That brings up two more points. Deviating from the
system and not setting any target. First anyone using a system for targets and
stops must thoroughly understand the system and be comfortable with it for it to
work. If you understand it you are more likely to stick with it most of the
time. When you do deviate, you only do so when it makes sense and with a full
understanding of the risks and potential corrective actions if wrong.</P>
<P> There are times when I feel so strongly about
letting my profits run that I won’t set any target. This is usually when a
stock has broken out to a new all time high on good momentum. In that case I
just set the stop as described above and hope it won’t get hit for a long
time. Years is fine <G>. Some would say that it never makes sense to set
targets, but I disagree with that. My experience is that reasonably thought out
targets pay off.</P>
<P> My closing point is that targets should be set as
limit orders with your broker but stops should always be mental stops. The
reason is that if you have a day job like I do, you can't watch the market all
the time. You want your position to be closed when the target is hit, so a limit
order is called for. Now to keep down the number of limit orders, I never submit
them until my stock gets within reasonable distance of the target. The reason
that stops should be mental is that I firmly believe that specialist do gun
stops. They like to take a stock down to shake out the stops just before
allowing it to run. Therefore my stop is mental and I require the stock to close
below the stop before closing the position the next morning. The cost of this is
that you can get burned if a stock is dropping fast, but I believe you save more
in the long run by avoiding shakeouts. I like to think I’m fairly rigid in
this except I do, as in the case with MSFT, take the next mornings market action
into consideration. Again you need to thoroughly understand what you are doing
when/if you violate the system.</P>
<P> That’s it, any comments pro or con? Does
anyone disagree with setting targets or stops? If so, why? What methods do you
use and why?</P>
<P>JimG </P></DIV></BODY></HTML>
</x-html>From ???@??? Sat Dec 05 13:14:23 1998
Received: from listserv.equis.com (204.246.137.2)
by mail02.rapidsite.net (RS ver 1.0.2) with SMTP id 2649
for <neal@xxxxxxxxxxxxx>; Sat, 5 Dec 1998 16:13:24 -0500 (EST)
Received: (from majordom@xxxxxxxxx)
by listserv.equis.com (8.8.7/8.8.7) id NAA21249
for metastock-outgoing; Sat, 5 Dec 1998 13:26:03 -0700
X-Authentication-Warning: listserv.equis.com: majordom set sender to owner-metastock@xxxxxxxxxxxxx using -f
Received: from freeze.metastock.com (freeze.metastock.com [204.246.137.5])
by listserv.equis.com (8.8.7/8.8.7) with ESMTP id NAA21246
for <metastock@xxxxxxxxxxxxxxxxxx>; Sat, 5 Dec 1998 13:26:00 -0700
Received: from mail-q1.pcmagic.net (mail-q1.pcmagic.net [206.117.211.8])
by freeze.metastock.com (8.8.5/8.8.5) with SMTP id NAA29095
for <metastock@xxxxxxxxxxxxx>; Sat, 5 Dec 1998 13:37:04 -0700 (MST)
Received: (qmail 2893 invoked by uid 3301); 5 Dec 1998 12:26:39 -0800
Received: from mail.pcmagic.net (206.117.211.6)
by mail-q1.pcmagic.net with SMTP; 5 Dec 1998 12:26:39 -0800
Received: from default ([206.117.27.20]) by mail.pcmagic.net
(Netscape Mail Server v2.02) with SMTP id AAA225
for <metastock@xxxxxxxxxxxxx>; Sat, 5 Dec 1998 12:26:38 -0800
Message-ID: <199812051226250750.06E8D436@xxxxxxxxxxx>
In-Reply-To: <Pine.3.89.9812050954.A7645-0100000@xxxxxxx>
References: <Pine.3.89.9812050954.A7645-0100000@xxxxxxx>
X-Mailer: Calypso Lite Version 2.40.40
Date: Sat, 05 Dec 1998 12:26:25 -0800
From: "Don Hughes" <daringdon@xxxxxxxxxxx>
To: metastock@xxxxxxxxxxxxx
Subject: Re: Moderated List
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by listserv.equis.com id NAA21247
Sender: owner-metastock@xxxxxxxxxxxxx
Precedence: bulk
Reply-To: metastock@xxxxxxxxxxxxx
X-Loop-Detect: 1
X-UIDL: 35e883ec3668da79f8d7129a9a04d33f
Jim:
I went to the link shown below, but could not find you. Can you explain in more detail?
Don Hughes
daringdon@xxxxxxxxxxx
*********** REPLY SEPARATOR ***********
On 12/5/98, at 9:43 AM, Jim Michael wrote:
>I think interest in the moderated list waned as the political BS ceased
>here. There has been little traffic and I am not happy with the
>Majordomo configuration supplied by my service providor (such as a
>dump of all previous posts being resent to everyone on the list) so I
>will be officially closing it.
>
>I have created a new TA discussion forum on dejanews. It is web based
>so there should be no problem with the posting of grahics or the use of HTML
>or XML. See the link at http://xmlworks.com/quotemonster/
>
>Cheers,
>
>Jim
>
>PS: A new script is up on the website. PseudoRT converts tick data for
>use in your EOD charting program. http://xmlworks.com/pseudort/
>
>On Sat, 5 Dec 1998, Staffan wrote:
>
>> What happened to the "Moderated List" ?
>> I once recieved a "Welcome Message" - but never any mails...
>> Even worse - I lost the e-mail address.
>> Does anyone know?
>> Regards,
>> Staffan
>>
>>
>>
>>
|