Fit: how to return chi-squared instead of r-squared

7 次查看（过去 30 天）
Gabriele
Gabriele 2014-1-22
回答： Bruno Pop-Stefanov 2014-1-22
Hi, i'm performing this fit:
f=fittype('a*(x^(-b))*exp(-x/c)');
options = fitoptions('Method','NonlinearLeastSquares','StartPoint',[100,1,1]);
[val1,val2,val3]=fit(x,y,f,options);
disp(val2);
and I get on the screen
sse: 9.7986e-04
rsquare: 1.0000
dfe: 97
adjrsquare: 1.0000
rmse: 0.0032
How could I get chi-square instead of r-square?

请先登录，再回答此问题。

回答（1 个）

Bruno Pop-Stefanov
Bruno Pop-Stefanov 2014-1-22
The Curve Fitting Toolbox only supports these goodness-of-fit statistics for parametric models:
If you want the chi-squared statistic, you will have to write your own code from your input data and your calculated fit.

类别

在 Help Center 和 File Exchange 中查找有关 Fit Postprocessing 的更多信息

标签

产品

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by