Regression with custom interception

[JP-Reis]

Hello community,

I have been trying to figure out how to obtain the regression equation for a linear regression model with a pre-defined intercept value with no success.

The customer want to replicate some of the Excel functionalities that they are used to use in Excel. Is this even possible in with this lib?

Thank you.

[JP-Reis]

I know that there is a built-in function to FitThroughOrigin, but what if I want to model the equation to intercept (a=xyz) on Y=3?

[diluculo]

You can apply y - 3.0 to the FitThroughOrigin function to model the equation with an intercept of 3 on the y-axis.

[JP-Reis]

Ok, so far that seems to do the trick!

Is there any similar function as FitThroughOrigin for exponential and polynomial models?

[diluculo]

var slope = FitThroughOrigin(x, y.Select(p => p - 3.0).ToArray())

[diluculo]

Please see mathnet-numerics/src/Numerics/Fit.cs.

For general models, nonlinear least square fitting (Levenberg-Marquardt) or other optimization methods can be used. Please see the test cases in mathnet-numerics/src/Numerics.Tests/OptimizationTests.

[JP-Reis]

I'll give it a look. Thank you.