How can I compute coefficient of determination(R^2) of robust regresstion?
2 views (last 30 days)
Show older comments
x=[1,2,3,4,5,6,7,8,9,10];
y=[1.1,2.1,13.1,4.3,5.4,5.8,6.9,7.8,16,10.1];
[b,stat]=robustfit(x,y,'bisquare');
sse=stat.dfe*stat.robust_s^2;
phat = b(1) + b(2)*x;
ssr = norm(phat-mean(y))^2; % ssr = norm(phat-mean(phat))^2; are also tested
R^2=1-sse/ssr
but sse and ssr are not same to the results of cftool ,nor the R^2.
thak you for your help
0 Comments
Answers (0)
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!