How to fit a curve with forced initial value

Dear everybody,

I'm trying to fit a serie of datapoints with a function f(x) (e.g. a polynomial function)by forcing the value of the function at x=0. I tried to find the solution using "fit" and "polyfit" Matlab functions but the interpolating algorithm doesn't seem to take into account the initial value that I want. Could you help me ? Thanks by advance.

Answer 1

[Richard Willey]

Hi Phillippe

If you have Statistics Toolbox or Curve Fitting Toolbox its relatively easy to force this type of constraint.

Here's a simple example using Curve Fitting Toolbox where I constrain the intercept to take the value "4" by setting the upper and lower bound for the regression coefficient to be 4.

X = 1:100
X = X'
Y = 3* X + 5 + randn(100,1)

foo = fit(X, Y, 'poly1', 'Lower', [-Inf 4], 'Upper', [Inf 4])