| Author |
 Topic  |
|
|
egonnelli
Brazil
9 Posts |
 Posted - 10/04/2012 : 5:50:01 PM
|
Hello,
I am in doubt about the "Parameters Initialization" from the "Holliday Model" Built Function. The question is understand this function and apply the same script for the "Parameters Initialization" to another equation.
The Holliday Model function is in the form: "y=1/a+b*x+c*x^2" and the script for initialization is:
------------------------------------------------------
sort(x_y_curve);
smooth(x_y_curve);
x_y_curve = 1.0 / x_y_curve;
double coeff[3];
fitpoly( x_y_curve, 2, coeff);
a = coeff[0];
b = coeff[1];
c = coeff[2];
-------------------------------------------------------
Now my question,
If I want to Apply this script for another equation, in this case:
"y = K/(a+b*x^2+c*x^4)+ T; " (It looks like the above equation, but substituting "1.0" by K. And changing the degree of the denominator polynomial for 4th. Plus a constant "T")
What is the new script in these conditions?
Thank you very much,
Eduardo |
|
|
egonnelli
Brazil
9 Posts |
Posted - 10/04/2012 : 8:21:17 PM
|
Another thing,
What is the logic in the Initialization Parameters script, because in the Holliday1 function, y=a/(a+b*x+c*x^2), the script is:
sort( x_y_curve );
smooth( x_y_curve );
x_y_curve = 1.0 / x_y_curve;
double coeff[3];
fitpoly( x_y_curve, 2, coeff);
a = 1 / coeff[0];
b = a * coeff[1];
c = a * coeff[2];
Note that,in the case of Holliday function, y=1/(a+b*x+c*x^2), the script is simpler than the Holliday1
sort( x_y_curve );
smooth( x_y_curve );
x_y_curve = 1.0 / x_y_curve;
double coeff[3];
fitpoly( x_y_curve, 2, coeff);
a = coeff[0];
b = coeff[1];
c = coeff[2]; |
 |
|
| |
Topic |
|
|
|
Creating a new Parameters Initialization Script