How to fit a model a=bx+cy form, for series of data?

Question

Annu Panwar 2017-6-23

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/345959-how-to-fit-a-model-a-bx-cy-form-for-series-of-data

回答： Bernhard Suhm 2017-12-6

a=[3, 4, 6, 7,8] b=[5,9,10,12,19] a=[3,4,5,6,7,9]

x and y are known how to get the value of x and y using linear fit? I can solve the equations and get the value but I want only one values for x and y. Please provide help?

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

Hendrik 2017-11-27

the last a should be c??? If yes what about the dimensions? dim(a & b) = [1,5]; dim(a'' =?c) = [1,6];

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Bernhard Suhm 2017-12-6

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/345959-how-to-fit-a-model-a-bx-cy-form-for-series-of-data#answer_294911

If you interpret a=bx+cy as a linear equation with the scalars x,y as unknown (even though you said they are known), it's an overconstrained system of linear equations and there is no solution - but you probably knew that.

If you interpret your vectors to specify a set of datapoints, with a as the 'response' and b and c as the predictors, you can force MATLAB to fit a linear model with tbl = table(c',b',a','VariableNames',{'c','b','a'}); lm = fitlm(tbl,'a~b+c','Intercept','false');

Note that I transposed your row vectors into columns to build the data table ("frame").