|
PureBytes Links
Trading Reference Links
|
These are good questions John.
I developed the majority of the formulas sometime ago and then added to them
over time - so its possible there is some duplication (re:the 40 week m.a.).
Anyways I just wanted you to know I received your email and I will work on a
response but it may take me a little time as my daughter gets most of time
during her waking hours which lately seem to run till 10:00pm. (A little late
for a 2 yr. old)
That last one on the price oscillator will take some time to think thru again.
The parameters on the std indicators represent in in my opinion when the
underlying equity is moving from weak or neutral into strong territory (i.e.
starting to show strength).
David Duggan
John Hunter wrote:
> David,
> Thank you for contribution. I am sorry to have not got back to you sooner
> but with two Aussies working their way to the final of the US Open and the
> Australian team picking up more gold at the Commonwealth Games than is held
> by our Reserve Bank it has been too easy to get distracted. Roll on Sydney
> and 2000.
> When we get something like your posting I like to pull it apart and try to
> understand what each part is meant to do. One first question I have is what
> is the significance of the 40 day moving average as you use it three times
> in the system, in fact you use (Mov(C,40,S),>,Ref(Mov(C,40,S),-1) in the
> formula as well as a filter in the Explorer (and a third time if you count
> its use in colB)?
> >>From what I can see the things you are trying to measure can be grouped as
> follows:
>
> 1. You test for when a number of standard indicators are above some value,
> eg
> * If(RSI(14),>,55,1,0) and
> * If(C,>,SAR(0.02,0.20),1,0) and
> * If(MFI(14),>=,65,1,0) and so on.
>
> 2. You test for when a variable is greater today than when it was yesterday,
> eg
> * If(TRIX(12),>,Ref(TRIX(12),-1),1,0) and
> * If(Mov(EMV(14,S),3,S),>,Ref(Mov(EMV(14,S),3,S),-1),1,0) and
> * If(Mov(C,40,S),>,Ref(Mov(C,40,S),-1),1,0) and so on.
>
> 3. Things being greater than their moving average, eg
> * If(MACD(),>,Mov(MACD(),9,E),1,0) and
> * If(OscP(10,20,E,$),>,Mov(OscP(10,20,E,$),10,S),1,0) and so on.
>
> 4. Some "mix up" type measures, like:
> * If(((H+L+C)/3),>,(Mov(C,20,S)+(2*Std(C,20))),1,0) and
> * If(Mov(C,40,E)-Mov(O,40,E),>,0,1,0) or even
> *
> If(Mov(C,10,S),>=,Ref(Mov(C,10,S),-1),If(Mov(C,20,S),>=,Ref(Mov(C,20,S),-1),
> If(Mov(C,10,S),>=,Mov(C,20,S),1,0),0),0) (which is, by the way a great set
> of nested "ifs", but I am still working on the logic involved here requiring
> all three conditions to be true to get a single point. Sure it reduces the
> weighting you assign to moving averages of the price but seeing that it
> clearly defines an uptrend why not give it a higher weighing?).
>
> 5. Mysteries, like:
> *
> If(Abs(Sub(OscP(10,20,E,$),Mov(OscP(10,20,E,$),10,S))),>=,Ref(Abs(Sub(OscP(1
> 0,20,E,$),Mov(OscP(10,20,E,$),10,S))),-1),1,0) and
> * If(Mov(C-OscP(1,31,E,$),2,E),>,Ref(Mov(C-OscP(1,31,E,$),2,E),-1),1,0).
> Simple on the face of it requiring a value today to be greater than it was
> yesterday but what is the thing that produces the value doing? Both of these
> are complete mysteries to me. What can you tell us about them, particularly
> the first. I will start on these by turning them into functions and plotting
> them against the price chart just to get an idea of what they are meant to
> show.
>
> A final note is that I generally trade futures so your suggested explorer is
> of no use to me but I have worked on a simple system that uses the DDTI and
> a moving average of it to generate buy and sell signals on the long side
> only of the All Ords Index (our DJIA) with some success ($10,000 to $90,000
> in three years so it has enough start there to maintain interest).
>
> Will keep you informed of progress offline as time permits.
>
> JH
>
> -----Original Message-----
> From: David Duggan <mmaker@xxxxxxxxxxx>
> To: metastock@xxxxxxxxxxxxx <metastock@xxxxxxxxxxxxx>
> Date: Friday, 11 September 1998 2:53
> Subject: Re: "Trade The Trend" - Freebie - DD Trend Formula
>
> >Here is a freebie:
> >
> >I will call it the DD Trend Formula for now (MS 6.0 is a minimum reqm't).
> Its
> >a trend following formula that will avoid many (maybe not all) whipsaws.
> (I
> >really want to test this on thousands of securities to see what happens in
> the
> >long run.)
> ....................snip.............................snip...................
> ......
> >Again any and all comments welcome!!!
> >
> >David Duggan
|