|
PureBytes Links
Trading Reference Links
|
<FONT face=Arial color=#0000ff
size=2>JO,
What I
was asking for though was the description of the signals in your own
words.
<FONT face=Arial color=#0000ff
size=2>
For
example:
<FONT face=Arial color=#0000ff
size=2>
<FONT face=Arial color=#0000ff
size=2>You go long on the open of then next day
when 1) The close of the $VLE is 4% or more than the
prior close of the $VLE or 2) the close of the SPY crosses above the
20 day simple moving average of the closes, which ever occurs
first.
<FONT face=Arial color=#0000ff
size=2>
<FONT face=Arial color=#0000ff
size=2>You go short on the open of then next day
when 1) The close of the $VLE is 4% or below the
prior close of the $VLE or 2) the close of the SPY crosses below the
20 day simple moving average of the closes, which ever occurs
first.
Is this it or have I missed
something?
Joe J.<SPAN
id=hbblock>
<FONT face=Tahoma
size=2>-----Original Message-----From: manohohman
[mailto:no_reply@xxxxxxxxxxxxxxx]Sent: Thursday, November 20, 2003
9:49 AMTo: equismetastock@xxxxxxxxxxxxxxxSubject: Re:
[EquisMetaStock Group] SPY code updateIn message 9729
I posted the code for just the VLE portion and not the added 20 dma. Using
this code you don't have to have the VLE on the screen. When you attach it
to the SPY, it works just like it's suppose to. That's not the
problem.The problem is when it's plugged into the system's tester the
MS results are much different from the results found in other systems
testers. The difference between the results is too great to be
accounted for by algorithm differences alone. In the next few
days, I'm going to examine the code, the theory, the back testing methods,
the indices values from Esignal and whatever else I can find. This all
takes a lot of time so I don't know when I'll finish it. I'm suppose to be
finishing up a magazine article and I'm behind on that so I'll have to
juggle my schedule. Since it's not that important, I'll probably let this
slide a little for now. There are reams of test data on this index and
this method. It shouldn't be a big deal, but it's turning out to be one.
You can use the trend following system with or without the 20 dma.
The 20 dma improves the ability to use it as a trading system. You get
a lot of small losers but you catch the big trends where the money is.
Feel free to play with the code. It's there for whatever purpose you
can find useful. Roy tweaked part of it again for prev statements.
It's as tight as it's going to get in that regard. Since this is only
applied to the SPY EOD, whichever version of the base code you use
won't slow the processor down much. Nanoseconds. Have fun! If I
ever figure this out, I'll post it. I've got a version of this for the
Qs that works really well, but given the difficulty I have had with this
one that's generally well understood and tested to death, I don't think
I'll post the one for the Qs. JO--- In
equismetastock@xxxxxxxxxxxxxxx, "Joe J." <jojab@xxxx> wrote:> Hi
JO,> > What are the specific rules you are trying to code
for your buys and sells? If you provide a general description
perhaps I (or someone else) can give it a shot on 8.01. I know you
mentioned a 4% rule on the $VLE and a 20dma rule but what are they
specifically?> > Good Trading,> > Joe
J. > > > > >
> -----Original Message-----> From: manohohman
[mailto:no_reply@xxxxxxxxxxxxxxx]> Sent: Wednesday, November 19, 2003
11:23 PM> To: equismetastock@xxxxxxxxxxxxxxx> Subject:
[EquisMetaStock Group] SPY code update> > > Thanks
Martin. I appreciate your assistance. Unfortunately we're > still
stuck.> > Here's the code for the VLE part. I wrote it as an
expert so you can > attach it directly to the SPY chart. However,
it does not test out > properly. I'm looking at the formula and MS to
see if either one is > the problem. > > To make an
expert> > Highlight Buy> >
a:=Security("online:$VLE",C);> X:=If(Cum(Ref(a>Lowest(a)*1.04 OR
a<Highest(a)*.96,-1))=0,>
If(a>Lowest(a)*1.04,1,>
If(a<Highest(a)*.96,-1,0)),> If(PREV=1,>
If(a<HighestSince(1,PREV<1,a)*.96,-1,PREV),>
If(a>LowestSince(1,PREV>-1,a)*1.04,1,PREV)));>
I:=Cum(1)<=2;> State:=BarsSince(X=1 OR I)<BarsSince(X=-1 OR
I)+(Cum(X=1)=1);> State;> > Symbols Buy> >
a:=Security("online:$VLE",C);> X:=If(Cum(Ref(a>Lowest(a)*1.04 OR
a<Highest(a)*.96,-1))=0,>
If(a>Lowest(a)*1.04,1,>
If(a<Highest(a)*.96,-1,0)),> If(PREV=1,>
If(a<HighestSince(1,PREV<1,a)*.96,-1,PREV),>
If(a>LowestSince(1,PREV>-1,a)*1.04,1,PREV)));>
I:=Cum(1)<=2;> x=1 AND Ref(x,-1)=-1> > Symbols
Sell> > a:=Security("online:$VLE",C);>
X:=If(Cum(Ref(a>Lowest(a)*1.04 OR a<Highest(a)*.96,-1))=0,>
If(a>Lowest(a)*1.04,1,>
If(a<Highest(a)*.96,-1,0)),> If(PREV=1,>
If(a<HighestSince(1,PREV<1,a)*.96,-1,PREV),>
If(a>LowestSince(1,PREV>-1,a)*1.04,1,PREV)));>
I:=Cum(1)<=2;> x=-1 AND Ref(x,-1)=1> > You don't need
a highlight sell if you color your line red. It's the > inverse.
> > This is without the 20 dma. You can add that to the chart by
hand.> > Everything looks right but something is out of wack.
Back to the > drawing board. it predicts the trend changes nicely.
> > > > > > > --- In
equismetastock@xxxxxxxxxxxxxxx, "praktikus_ms" > <praktikus@xxxx>
wrote:> > Hi JO,> > > > What people like you,
Roy and many others do here in this group is > > greatly
appreciated by me (and I guess many others too). It helps > me >
> a lot to get a greater unterstanding of MS. Thank you all for your
> > posting!> > > > A really good way of further
improvement is the give and take rule. > > Only the one who
teaches realizes what he can learn from that.> > > > JO, I
was not sure wich version of MS you are using and from my > >
expirience I know that a lot of people dont know the Security Data
> > function. > > > > I was just trying to
assist with my little knowledge and was not > > aware of the
problems you described. From my use I did not come up > > with
similar problems.> > > > Thank you,> > >
> Martin > > > > > > > While this
one looks the same on the surface, Roy and I have been > > >
exchanging data and it has yet to work properly. Roy has found at >
> > least part of the problem with false signals and overlapping
> > signals > > > leading to wrong interpretations of
trades. Hopefully in a couple > > of > > > more
days, we'll have it ironed out. The backtesting seems to > work
> > > better for Roy using MS 7.02. MS 8.01 still has a few
issues. As > I > > > said these problems didn't come
up on TradeStation. MS is a > little > > > more
ridgid.> > > > Yahoo! Groups
Sponsor > >
ADVERTISEMENT> <<A
href="">http://rd.yahoo.com/SIG=12cfr7dsk/M=267637.4116732.5333197.1261774/D=egroupweb/S=1705375617:HM/EXP=1069395766/A=1853618/R=0/*<A
href="">http://www.netflix.com/Default?mqso=60178338&partid=4116732>
click here > <<A
href="">http://us.adserver.yahoo.com/l?M=267637.4116732.5333197.1261774/D=egroupmail/S=:HM/A=1853618/rand=221272576>
> > To unsubscribe from this
group, send an email to:>
equismetastock-unsubscribe@xxxxxxxxxxxxxxx> > > >
Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service <<A
href="">http://docs.yahoo.com/info/terms/>
.To
unsubscribe from this group, send an email
to:equismetastock-unsubscribe@xxxxxxxxxxxxxxxYour
use of Yahoo! Groups is subject to the <A
href="">Yahoo! Terms of Service.
Yahoo! Groups Sponsor
ADVERTISEMENT
To unsubscribe from this group, send an email to:
equismetastock-unsubscribe@xxxxxxxxxxxxxxx
Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.
|