Try using or tightening upper and lower bounds on coefficients.

Question

ali alsabari 2019-8-25

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/477510-try-using-or-tightening-upper-and-lower-bounds-on-coefficients

回答： Bjorn Gustavsson 2019-8-26

95_data.xlsx

hello

please i need your help the attatched file is my data and i want to do matlab curve fittining using custom equation:

4.112-k1*exp(-[x-477/m])-k2*exp(-[x-477/d])

i got error as in the title the values in the equation based on knowledge from data itself as paramters that already know and i want to know the other coffecient of the equation .. i have tried many methods and reducing using different algorithm using command but i still get an error please help me with this

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

Matt J 2019-8-25

Please refer to this.

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Bjorn Gustavsson 2019-8-26

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/477510-try-using-or-tightening-upper-and-lower-bounds-on-coefficients#answer_389045

For arbitrary fitting tasks I personally prefer to use weighted least-square minimization using some optimization function,

if lsqnonlin doesn't work for you you could try the file exchange contributions: fminsearchbnd, minimize. That way I have more direct control of what is done.

HTH