MATLAB: Weighted least square fit

curve fittingCurve Fitting Toolboxleast-squareMATLABweighted least square

I have a set of data points. I want to fit a straight line so that the least square error is minimized. But different points have different weighting factors. Is there any built-in function for doing this?

-saima

Best Answer

Yes. There is LSCOV.

Related Solutions

MATLAB: How to display the weighted residuals when fitting data with weights in Curve Fitting Toolbox 1.2.1 (R2008a)

The ability to display residuals normalized by the weights in weighted least squares curve fitting is not available in the curve fitting tool in Curve Fitting Toolbox 1.2.1 (R2008a).

To workaround this issue, you could generate a MATLAB file from the curve fitting tool and modify the code to display weighted residuals. Alternatively, you could also use the programmatic curve fitting interface.

MATLAB: Weighted Least Squares fit

You can certainly make visual adjustments to the weights to get a better fit - but what would the fit mean? I am assuming that there is some purpose to fitting the data besides making a nice plot.

When you're doing least squares fits, you're fitting the data to some statistical model. The most common model is that the errors in your measurements are independent and have equal variance. If they don't have equal variance, then you use weighted least squares. If you don't know the variances, there are methods for estimating them. See this earlier discussion.

Best Answer

Related Solutions

MATLAB: How to display the weighted residuals when fitting data with weights in Curve Fitting Toolbox 1.2.1 (R2008a)

MATLAB: Weighted Least Squares fit

Related Question