Well, outside of that I think most of the other obvious ideas have been covered in the various adaptive MA's that others have developed. But I'll keep thinking Steve...
As an aside, where can I find Tomasz's TASC stuff...? Thanks Rob --- In [email protected], "Steve Dugas" <[EMAIL PROTECTED]> wrote: > > Hi Rob - Yes, I am familiar with Ehlers stuff ( well, at least to the > limited extent that I am smart enough to understand it, but fortunately I > can translate his code without having to understand all the little details > 8 - ). I have already translated a bunch of his stuff to AFL and there is > also quite a collection in the AFL library and in TJ's TASC formulas. I have > to admit it looks pretty high-tech and sexy but I was never really able to > make a ton of money with it and I have moved on now... > > For the sake of argument, let's just say I am trand following since I think > the basic rationale behind what I am doing is pretty similar....If anything > else should come to mind, please let me know...Thanks for your ideas! > > Steve > > ----- Original Message ----- > From: "sidhartha70" <[EMAIL PROTECTED]> > To: <[email protected]> > Sent: Monday, June 09, 2008 5:34 PM > Subject: [amibroker] Re: Creating "period" arrays for built-in averages > > > > Well Steve, again it all depends on how you are trading... that kind > > of defines what makes sense as a 'period' if you like. > > > > The first thing that popped into my mind when I read your post was > > using something like John Ehler's digital signal processing techniques > > to define somekind of 'dominant market cycle period'... > > > > Are you familiar with Ehler's stuff...?? www.mesasoftware.com > > John's pretty open about the techniques he uses and buying his > > software is really only for the 'bone idle'... you can work it all out > > yourself. He lists EasyLanguge code to calculate dominant cycle > > periods in a couple of his books... 'rocket science for traders' and ' > > cybernetic analysis for stocks and futures'... > > > > Do send me a message if you want or need further info. > > > > Yours > > > > Rob > > > > --- In [email protected], "Steve Dugas" <sjdugas@> wrote: > >> > >> Hi - I am not trend following in the popular sense but I am trying > > to smooth > >> another array that contributes to my signals in order to make it less > >> volatile. Right now I am just experimenting with some different > > averages to > >> determine which might work best, including some of the popular AMA's > > like > >> VIDYA, KAMA, MAMA etc, and it dawned on me that I might be able to > > easily > >> create a bunch of different AMA's to test by just starting with > > different > >> momo/volatility oscillators and then converting these arrays to > > periods that > >> are fed into AB's assortment of built-in MA's. Except I am not sure > > how to > >> best convert the values of the different oscillators to numbers that > > would > >> work well as "periods". Ideally I would like to be able to optimize > > also so > >> I am thinking that maybe my conversion formula should include some > > sort of > >> smoothing constant whose value I can optimize on. Right now I just > > thought > >> of it and it is all swimming around in my head so I was hoping > > someone might > >> be able to point me in the right direction. Do you think you might > > be able > >> to say a few words about some of the different techniques? I could > > probably > >> fill in most of the details by googling if I had a better idea of the > >> possible alternatives or starting points...Thank very much! > >> > >> Steve > >> > >> ----- Original Message ----- > >> From: "sidhartha70" <sidhartha70@> > >> To: <[email protected]> > >> Sent: Monday, June 09, 2008 3:31 PM > >> Subject: [amibroker] Re: Creating "period" arrays for built-in averages > >> > >> > >> > Steve, > >> > > >> > There are so many techniques you could use to create a variable array > >> > for use in MA calculations... I think it all depends on 'how' you want > >> > to use the MA. Clearly the technique needs to make some intuitive > >> > sense. Are you trend following and want to avoid whipshaws in volatile > >> > markets....? > >> > > >> > > >> > > >> > --- In [email protected], "Steve Dugas" <sjdugas@> wrote: > >> >> > >> >> Hi - I see that most of AB's built-in moving averages will take an > >> > array for the Periods arg. I imagine a typical use of this might be to > >> > calculate some array based on momentum, volatility, whatever, then > >> > convert that to an array representing suitable periods that these > >> > functions can use? There are many indicators/oscillators etc that try > >> > to measure momo/volatility - my question is, what would be some good > >> > measures to use, and then some good methods of converting these > >> > different arrays to an array of periods suitable for use in MA(), > >> > DEMA(), etc? Are there other/better ways to arrive at an array of > >> > "periods"? Of course if anyone is doing this and has an example they > >> > could post, I would be very grateful. Thanks for any help! > >> >> > >> >> Steve > >> >> > >> > > >> > > >> > > >> > ------------------------------------ > >> > > >> > Please note that this group is for discussion between users only. > >> > > >> > To get support from AmiBroker please send an e-mail directly to > >> > SUPPORT {at} amibroker.com > >> > > >> > For NEW RELEASE ANNOUNCEMENTS and other news always check DEVLOG: > >> > http://www.amibroker.com/devlog/ > >> > > >> > For other support material please check also: > >> > http://www.amibroker.com/support.html > >> > Yahoo! Groups Links > >> > > >> > > >> > > >> > > >> > > > > > > > > ------------------------------------ > > > > Please note that this group is for discussion between users only. > > > > To get support from AmiBroker please send an e-mail directly to > > SUPPORT {at} amibroker.com > > > > For NEW RELEASE ANNOUNCEMENTS and other news always check DEVLOG: > > http://www.amibroker.com/devlog/ > > > > For other support material please check also: > > http://www.amibroker.com/support.html > > Yahoo! Groups Links > > > > > > > > >
