Fit a square root function to data
147 views (last 30 days)
Show older comments
I would like to fit a function of form y=K*x^.5+c, where Matlab finds the best fit values of K&c. What's the best way to do this? Thanks!
0 Comments
Answers (1)
Star Strider
on 15 Sep 2014
Edited: Star Strider
on 15 Sep 2014
Your function is actually linear, so you can use any linear regression function such as the Statistics Toolbox regress function, since it supplies several statistics on the fit.
Otherwise, use the mldivide function or ‘\’ operator. Assuming x and y are row vectors in your original data:
x = linspace(0, 10, 15); % Create Data
y = 3.*sqrt(x)+5 + 0.1*randn(size(x)); % Create Data
p = [sqrt(x)' ones(size(y'))]\y'; % Estimate Parameters
The vector of estimated parameters correspond to p(1)=K and p(2)=c.
If x and y are column vectors, eliminate the transpose (') operators in the ‘p’ calcualation.
You can also use polyfit and its friends, remembering to use sqrt(x) instead of x in the argument list:
yp = polyfit(sqrt(x), y, 1);
0 Comments
See Also
Categories
Find more on Linear and Nonlinear Regression in Help Center and File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!