[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Dec 2001 cotton [tomasz]



PureBytes Links

Trading Reference Links

Dear Tomasz;

Thank you VERY much for the exploration formula. Could this formula be 
expanded upon using points or percentage difference of previous 
close, whether ( > or <).

I posted this study several days ago and you pointed me to Dimitri and 
several of posts that he had submitted. 

I assumed that Along with your duties as the Author of this great 
program (AMIBROKER), and I mean Great, That you were busy increasing 
the power of the program and fullfilling the users wish lists.

I have used other software packages,( I will not mention the names, 
Out of respect for AMIBROKER, but this program is the best and easiest 
to use. I would personally like to help promote this product to 
others. I recommend this software to everyone I talk to.

I felt very comfortable corresponding with Dimitri, he is very 
efficient in AFL coding, I like the work that he has done to date.
Also, others in the group as well, submitted some ideas.

But it is to you that I owe the greatest appreciation, you are the 
coding GURU. Thank you very much. THE POWER OF AMIBROKER IS 
UNSTOPABLE!!. Keep up the Great work.

If I can help in any way, please let me know. 

Best Wishes,
Anthony



--- In amibroker@xxxx, "Tomasz Janeczko" <amibroker@xxxx> wrote:
> Dear Anthony,
> 
> 
> Here comes the exploration formula that calculates the numbers you 
want:
> 
> ///////////////////////////////////////
> 
> monday = dayofweek() == 1;
> 
> // monday up after friday > thursday
> mup = monday AND ref( close, -1 ) < close AND ref( close, -2 ) < 
ref( close, -1 );
> 
> // monday down after friday > thursday
> mdown = monday AND ref( close, -1 ) < close AND ref( close, -2 ) < 
ref( close, -1 );
> 
> // monday up after friday < thursday
> mup2 = monday AND ref( close, -1 ) < close AND ref( close, -2 ) > 
ref( close, -1 );
> 
> // monday up after friday < thursday
> mdown2 = monday AND ref( close, -1 ) < close AND ref( close, -2 ) > 
ref( close, -1 );
> 
> totmup = lastvalue( cum( mup ) );
> totmdown = lastvalue( cum( mdown ) );
> totmup2 = lastvalue( cum( mup2 ) );
> totmdown2 = lastvalue( cum( mdown2 ) );
> 
> filter = totmup > 0;
> 
> numcolumns = 4;
> column0 = totmup;
> column1 = totmdown;
> column2 = totmup2;
> column3 = totmdown2;
> 
> //////////////////////////////////////////////////
> 
> Please switch to "Range: n last days" and enter 1 into the "n" 
field.
> Then click Automatic Analysis "Exploration" button.
> 
> Hope this one helps you.
> 
> One hint for the future: don't belive people saying something is 
impossible.
> If you need AFL guidance ask me.
> 
> Best regards,
> Tomasz Janeczko
> ===============
> AmiBroker - the comprehensive share manager.
> http://www.amibroker.com
> 
> 
> ----- Original Message -----
> From: <ajf1111@xxxx>
> To: <amibroker@xxxx>
> Sent: Saturday, July 28, 2001 6:35 AM
> Subject: [amibroker] Re: Dec 2001 cotton
> 
> 
> > Bill,
> >
> > Thank you for your link and response, yes I have done it in excel,
> > but, My spreadsheets produce numbers only, I do not produce the
> > probability cone as shown at the link address.
> >
> > I am just getting familiar with AFL code, If you know how to 
produce
> > the probability cone in AFL code as shown at the link address, 
please
> > submit.
> >
> > Also, If you know how to produce the study / exploration as 
previously
> > described, your help would be greatly appreciated.
> >
> > also, thank you for the link to the pattern software.
> >
> > best regards
> > Anthony
> >
> > --- In amibroker@xxxx, "wavemechanic" <wd78@xxxx> wrote:
> > > Anthony:
> > >
> > > If you can calculate the probability with Excel, why can't you 
do so
> > with AFL? Is your study similar to Fishback's ODDS probability 
cones,
> > which is summarized below?
> > > "ODDS Probability Cones (which are greatly influenced by recent
> > price volatility) provide you with a visual guide to the most
> > probable range of future prices. This range (i.e. the cone's 
width) is
> > determined by recent volatility in prices, the number of time 
periods
> > projected, and the probability percentage (e.g., 68% confidence, 
90%
> > confidence, etc.). The more volatile the security prices, the 
wider
> > the expected range of future prices and hence the wider the cones. 
The
> > cones always widen from the apex even if recent volatility is very
> > low, because as time increases, the better the odds of a 
significant
> > price move." From
> > 
http://www.paritech.com/education/technical/indicators/volatility/odds
> > .asp.
> > >
> > > Bill
> > > ----- Original Message -----
> > > From: ajf1111@xxxx
> > > To: amibroker@xxxx
> > > Sent: Friday, July 27, 2001 9:25 PM
> > > Subject: [amibroker] re: Dec 2001 cotton
> > >
> > >
> > > Dimitri;
> > >
> > > Thank you for your response to explorations. The study that I 
am
> > > trying to acheive can be used for stocks as well as 
commodities.
> > >
> > > Along with a visual confirmation as to bullish or bearish a
> > market, I
> > > also try to quantify market direction with several indicators
> > > confirming.
> > >
> > > To that end, one of my confirming studies deals with
> > probabilities.
> > > I have created several spreadsheets in microsoft excel 97, 
which
> > allow
> > > me to find the standard deviation, Historical Volatility, and
> > > Probability of market movement. By using these measurements 
and
> > time,
> > > I am able to determine what is the Probability of a target 
price
> > to be
> > > acheived or not. ( ex. 60% ,70%, 80%, 90%).
> > >
> > > If you would like me to upload a copy of this spreadsheet. Let 
me
> > > Know.
> > >
> > > Back to the exploration. I guess amibroker can not handle this
> > type of
> > > study.
> > >
> > > A sample report of this type of study might look like this;
> > >
> > > testing data: 90 days of history
> > > test 1: If Friday's close > Thursday's close
> > > Monday's close: higher-- 11
> > > lower -- 1
> > > test 2: If Friday's close < Thursday's close
> > > Monday's close: higher-- 11
> > > lower -- 1
> > > test 3: If Friday's close = Thursday's close
> > > Monday's close: higher-- 11
> > > lower -- 1
> > >
> > > As you can see with this type of test, there is a 90% chance 
of
> > > Monday's close being higher. The Probability is 90%, My visual
> > > confirmation (technical analysis) is bullish, My indicators 
are
> > > bullish, I will initiate a bullish strategy.
> > >
> > > If you Know of a program that can produce this type of study,
> > please
> > > let me know.
> > >
> > > Again, thank you,
> > > Anthony
> > >
> > >
> > >
> > > Yahoo! Groups Sponsor
> > > ADVERTISEMENT
> > >
> > >
> > >
> > >
> > > Your use of Yahoo! Groups is subject to the Yahoo! Terms of
> > Service.
> >
> >
> >
> >
> >
> > Your use of Yahoo! Groups is subject to 
http://docs.yahoo.com/info/terms/
> >
> >
> >