how to fit a curve with the given trend to the limited data points

1 次查看（过去 30 天）
Mos_bad
Mos_bad 2018-9-8
评论： Mos_bad 2018-9-9
Please find the attached.
I have plotted curve 2 with the limited data point and want to continue it up to the horizontal axis like curve 1. Therefore I used curve fitting. The trends of the curve 2 should be the same as curve 1 however, after fitting a curve (dashed line), it shows totally different behavior after endpoint of curve 2. please let me know how to continue fitting curve trend similar to curve 1. here is whatever I have written so far:
x1= linspace(min(Mu_s_srt1),10,200);
p=polyfit(Mu_s_srt1,(1-Mu_s_P1),2);
f1=polyval(p,x1);
hold on;
loglog(x1,f1,'--k','linewidth',2);
hold on;
loglog(Mu_s_srt1, 1-Mu_s_P1,'-b', 'linewidth',2);
grid minor;
hold on;
  7 个评论
显示 5更早的评论
Image Analyst
Image Analyst 2018-9-9
It's hard to force the curve down on the right side when you're extrapolating. If you don't have a point out there to help pull it down, and just go by what you have on the left, it might not go down there. How about if you just specify one point way out on the right to help pull it down. It doesn't have to be that accurate, just something to help pull the extrapolation downwards.
Mos_bad
Mos_bad 2018-9-9
That's correct, no need to be that accurate, just follow the red curve trend. You mean just adding some data points manually?

请先登录，再进行评论。

请先登录，再回答此问题。

回答（0 个）

类别

在 Help Center 和 File Exchange 中查找有关 Linear and Nonlinear Regression 的更多信息

标签

Community Treasure Hunt

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

Start Hunting!

Translated by