Get 95% Confidence Values from Fit Coefficients

Question

Scott 2011-12-19

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/24322-get-95-confidence-values-from-fit-coefficients

采纳的回答： Sean de Wolski

Hey, I've run a curve fit and have various coefficient outputs. For example:

A = 13.34 (12.37, 14.31)

B = 9.955 (6.182, 13.73)

Any idea how I can save a vector such as:

dA=(12.37, 14.31) and dB=(6.182, 13.73)? Everything is run via my .m file so I don't want to do this by any mouse-commands in a GUI but rather via a command within the .m file. Thanks.

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Sean de Wolski 2011-12-19

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/24322-get-95-confidence-values-from-fit-coefficients#answer_31922

在 MATLAB Online 中打开

per comments:

doc confint

will explain how to extract the confidence interval parameters from a fit object.

3 个评论
显示 1更早的评论隐藏 1更早的评论

Sean de Wolski 2011-12-19

Did you run the curve fit in the curve fit tool (cftool) of the curve fitting toolbox?

Scott 2011-12-19

I ran it via a command in the .m file since the fit is contained within a loop:

f = fittype(...)

[c2,gof2] = fit(xdata,ydata,f,...)

请先登录，再进行评论。

Answer 2

Scott 2011-12-19

1
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/24322-get-95-confidence-values-from-fit-coefficients#answer_31927

Sean's help worked! [c2,gof2] = fit(xdata,ydata...etc) ci = confint(c2) Gives the variable ci a vector of all the uncertainties. I can then save dA(1,1)=ci(1,1) and dA(1,2)=ci(1,2) etc.