Force intercept at (0,0) using polynomial regression with Apache Maths (Java)

Question

I am trying to fit a curve with a 2-nd degree polynomial regression using PolynomialCurveFitter in Java with Apache Maths. This is my code:

final WeightedObservedPoints obs = new WeightedObservedPoints();
obs.add(1, 3400);
obs.add(3, 6000);
obs.add(8, 9600);
obs.add(10, 30000);

// Instantiate a third-degree polynomial fitter.
final PolynomialCurveFitter fitter = PolynomialCurveFitter.create(2);


// Retrieve fitted parameters (coefficients of the polynomial function).
final double[] coeff = fitter.fit(obs.toList());
//System.out.println(coeff);
System.out.println("coef="+Arrays.toString(coeff));

And it's working fine, but I would like to intercept at (0,0), i.e. my constant term in the equation to be zero: y=ax+bx^2.

Thank you very much for your help, Fred

You can try to add: obs.add(0, 0); and change PolynomialCurveFitter.create(4); — axelclk
– axelclk, Commented Jul 8, 2015 at 16:23
I thought of adding obs.add(0, 0); but still it would not be as accurate as forcing it intercept (0,0). But why PolynomialCurveFitter.create(4);? — Federico
– Federico, Commented Jul 8, 2015 at 16:37

tim · Accepted Answer · 2016-12-02 01:42:42Z

2

You can add the point(0,0) with a huge weight, such as obs.add(10000, 0, 0). if you use obs.add(0, 0), the default weight is 1.

answered Dec 2, 2016 at 1:42

tim

214 bronze badges

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

Force intercept at (0,0) using polynomial regression with Apache Maths (Java)

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related