logo
Welcome Guest! To enable all features please Login or Register.

Notification

Icon
Error

Options
Go to last post Go to first unread
RobW  
#1 Posted : Saturday, October 27, 2007 9:59:03 AM(UTC)
RobW

Rank: Member

Groups: Registered, Registered Users, Subscribers
Joined: 10/27/2007(UTC)
Posts: 11

I have created Dr. Elder's Impulse System indicator in Metastock 10 but when I drag it to a chart all I get is a flat line with a couple of small peaks above or below the line in random spots.

The Impulse box says Impulse (0.0000, 0.0000).

Here is a pictuure of the results:

http://skunkwerkz.dyndns.org/charts/geye.htm

Here is what I have used as a formula posted on Metastocktools.com. Any help would be appreciated.

{ Dr Elder's Impulse system interpretation v1.1
Plot signals on own window below daily chart.
http://www.elder.com/MetaStock/Impulse.htm

http://www.metastocktools.com }

{ Variables user-input }
pdsNoEntry:=Input("No entry x periods from last exit",0,21,5);
pdsW:=Input("Weekly EMA periods",1,520,13)*5;
pdsD:=Input("Daily EMA periods",1,252,13);
pdsShort:=Input("Shorter MACD periods",
1,252,12);
pdsLong:=Input("Longer MACD periods",2,2520,26);
pdsSignal:=Input("MACD trigger signal periods",
2,252,9);
tradeDelay:=0;

{ Indicators }
EmaW:=Mov(C,pdsW,E);
EmaD:=Mov(C,pdsD,E);
Mac:=Mov(C,pdsShort,E)-Mov(C,pdsLong,E);
MacTrig:=Mov(Mac,pdsSignal,E);
Hist:=Mac-MacTrig;

{ System logic }
Out:={EmaW<Ref(EmaW,-1)
AND }EmaD<Ref(EmaD,-1)
AND Hist<Ref(Hist,-1);
In1:=EmaW>Ref(EmaW,-1)
AND EmaD>Ref(EmaD,-1)
AND Hist>Ref(Hist,-1);
In:=In1 AND BarsSince(Out)>pdsNoEntry;

{ System signals }
Init:=Cum(In+Out>-1)=1;
InInit:=Cum(In)=1;
flag:=Ref(BarsSince(Init OR In)
<BarsSince(Init OR Out)+InInit,-tradeDelay);
signals:=(InInit AND Alert(InInit=0,2)
OR flag AND Alert(flag=0,2))
-(flag=0 AND Alert(flag,2));

{ Plot }
0;signals

PTJim  
#2 Posted : Tuesday, October 30, 2007 12:03:10 AM(UTC)
PTJim

Rank: Advanced Member

Groups: Registered, Registered Users, Subscribers
Joined: 5/10/2006(UTC)
Posts: 252

Thanks: 11 times
Was thanked: 9 time(s) in 6 post(s)
Sorry I can't help you with debugging the code, but FYI the Elder plugin package for Metastock does an excellent job of implementing all his indicators from the books (including a good job coloring impulse bars), plus a number of explorations adhering to his methods (including MACD Histogram Divergence, Triple Screen and Fallen Angels) and some Expert Advisors. None of the above are password-protected; you can examine and modify any of them, although they do seem free of bugs as-is. Also FYI, you can get it from his elder.com website considerably cheaper than elsewhere.

I have no relation to Dr. Elder other than as a customer. If you want to just use his methods, perhaps modify them a bit, and do investor research rather than do all the coding work from scratch (and I understand the appeal of that to people who love to code) the Elder plugin is a solid value for the money. I doubt you could hire a Metastock programmer to create everything in it from scratch for 5 times the price.

RobW  
#3 Posted : Tuesday, October 30, 2007 6:08:15 PM(UTC)
RobW

Rank: Member

Groups: Registered, Registered Users, Subscribers
Joined: 10/27/2007(UTC)
Posts: 11

Thanks for the info. I knew the disk existed and I was not about to spend $299. When I saw it on his site for $199 it was a 'done deal' and the disk has been ordered.

Cheers

amory  
#4 Posted : Wednesday, October 31, 2007 5:57:05 AM(UTC)
amory

Rank: Advanced Member

Groups: Registered, Registered Users
Joined: 4/18/2007(UTC)
Posts: 99
Location: sydney australia

beginning to think that many of these magic formulas are overrated. tried it out on some charts, it works alright. the peaks above or below the line are simply buy or sell signals, the trouble is they don't prove much by way of prescience.

as I see it, Rainbow oscillator, Projection oscillator or just good old Stochastic work as well or better. just because a formula is complicated & written by someone famous doesn't necessarily mean a thing.

it so happens, I made a similar comment about SVAPO. hope I'm not being disruptive. if so, tell me & I shall be happy to desist. in the meantime I call it as I see it.

PTJim  
#5 Posted : Thursday, November 1, 2007 12:45:09 PM(UTC)
PTJim

Rank: Advanced Member

Groups: Registered, Registered Users, Subscribers
Joined: 5/10/2006(UTC)
Posts: 252

Thanks: 11 times
Was thanked: 9 time(s) in 6 post(s)
Amory, I agree with you to the extent that there are no "magic" formulas. While I regard some as just ridiculous hocus-pocus no better than casting chicken bones (Gann fans, Fibonnacci, Astrological cycles etc.), the rest are just tools.

There are only so many ways you can manipulate OHLCV data, after all, but if you're going to do TA you have to put together a toolbox you're comfortable with and understand. While there are "new" indicators popping up every day (RMO is my current favorite whipping-boy) I try to pay attention to those I can comprehend and which have a clearly-defined history and reason to be. In this respect I like Elder's, Bollinger's and Appel's work - they explain what their tools do, and WHY. Can anybody point to why Gann/Fibonnacci and especially lunar/astrological methods "work"?

And it sure seems no given indicator is sufficient; you need to package various tools to do different things (enter, set stops, exit) under different conditions (long, short, trending, sideways).


PTJim  
#6 Posted : Thursday, November 1, 2007 1:40:30 PM(UTC)
PTJim

Rank: Advanced Member

Groups: Registered, Registered Users, Subscribers
Joined: 5/10/2006(UTC)
Posts: 252

Thanks: 11 times
Was thanked: 9 time(s) in 6 post(s)
Hey Rob, just a quick followup.

After receiving a private message from Jose, I should make it clear that when I mentioned I couldn't debug your code, I wasn't implying there were any bugs in Jose's MS coding. In fact, I missed the fact that it *was* his code at first - read your posting too fast and just missed that. Jose informs me he believes it to be bug-free, so maybe it was something about the way you implemented it that bombed it on your system.

Also, I did find one very tiny problem in the Elder package, having to do with coloring a histogram. In one of his indicators (I can't remember which one at the moment, but think it was MACD Combo), the histogram bars are all one color (I think red) when below the zero line, whereas they *should* be both red and green depending on whether the histogram is rising or falling bar-over-bar.

This tiny bug is easily fixed by highlighting a problem bar (a red one that should be green or vice-versa), right-clicking and then selecting the correct color. After saving it, the problem goes away. As far as I know the calcs are fine - it's just the one instance of bar-coloring that's wrong.


amory  
#7 Posted : Thursday, November 1, 2007 4:03:26 PM(UTC)
amory

Rank: Advanced Member

Groups: Registered, Registered Users
Joined: 4/18/2007(UTC)
Posts: 99
Location: sydney australia

having tried them out on a variety of charts, come to the conclusion that Elders & SVAPO are not in the same class. Elders is actually very precise & predictive. the other one, well maybe I'm just not applying it correctly. the people who write these formulas know what they are doing, some more so than others.
Users browsing this topic
Guest (Hidden)
Forum Jump  
You cannot post new topics in this forum.
You cannot reply to topics in this forum.
You cannot delete your posts in this forum.
You cannot edit your posts in this forum.
You cannot create polls in this forum.
You cannot vote in polls in this forum.