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

Re: three line break DLL plug-in now available



PureBytes Links

Trading Reference Links

Barry, sorry for not responding earlier. didn't see your post until now.

before i answer you question, the line break method is a volatility
adaptive system. in other words, the reversal criteria changes
according to recent price movements. it is similar to point and
figure, except there is no fixed dollar reversal amount.

per your question, only variable 1 should be optimized. this is the
number of line breaks for a reversal. for three line break, it is 3.
the other thing is that variable 1 range should be from 2 to 5. i
have not programmed 1-line break. anything above 5 doesn't really make
sense. the second variable is a flag. only values of 4 for 'aligned'
lead line and 5 for 'aligned' lag line should be used. when using in
this type of system, use:

buy=cross(aceTLB(opt1,4),aceTLB(opt1,5));

version 1.1 has an additional variable, "aceTLBsl". this was added
because of a request. it shows at which price you would see a
potential reversal. to use it, the code would look like:

A=aceTLB(opt1,4);
buy=cross(aceTLBsl,C);

reply back with your results!

regards,
/j.
toronto.tasug.com

--- In amibroker@xxxx, "bdunne_uk" <barry@xxxx> wrote:
> Hi J,
> 
> This looks really useful. However, I have a query:
> 
> Before reading the pdf file, I set up an analysis that tried to 
> optimize everthing, even the flags. Here is the analysis code:
>
> Var1 = Optimize ("V1", 5, 1, 7, 1) ;
> Var2 = Optimize ("V2", 2, 1, 7, 1) ;
> Var3 = Optimize ("V3", 3, 1, 7, 1) ;
> 
> A1 = aceTLB(Var1,Var2);
> A2 = aceTLB(Var1,Var3);
> 
> Buy = Cross(A1,A2);
> Cover = Buy; /* Cover short when going long */
> Sell = Cross(A2,A1);
> Short = Sell; /* Open short when closing long */
> _____________________________________________________________________
> 
> The optimization produced 5,2,3 as shown above.
> 
> Then I realised that 2,3 only define how the chart is plotted (I 
> think), so I looked to see what 5,4,5 produced in the results. Guess 
> what, it is vastly inferior to 5,2,3.
> 
> Am I missing something here?
> 
> Regards, Barry.