sense of two term exponential model fitting

2 次查看（过去 30 天）
akle
akle 2016-12-10
评论： Image Analyst 2016-12-10
Hi. I have data in which Y axis decrease as the X is increasing. Now I want to fit it exponentially.
Firstly I tried single term exponential model. I get result as f(x) = a*exp(b*x)
Coefficients (with 95% confidence bounds):
a = 44.56 (42.77, 46.36)
b = -0.4552 (-0.4744, -0.4361)
here -0.4552 is my exponential slope, namely rate of decrease. However I am not satisfied with the figure due to bad fitting. Then I tried two-term exponential fitting, and i get:
f(x) = a*exp(b*x) + c*exp(d*x)
Coefficients (with 95% confidence bounds):
a = 71.51 (68.16, 74.86)
b = -0.66 (-0.6801, -0.64)
c = -8.108e+07 (-2.139e+08, 5.17e+07)
d = -9.655 (-10.63, -8.68)
In matlabs help it is said that if b and d both negative then the data tends to decrease. So nothing else.
In my case I have total 5 plots. for all plots the b and d values are as follows:
-0.6600 -9.6555
-0.5496 -15.4602
-5.3828 -0.7509
-1.3416 2.6037
-5.5994 -0.5741
I couldnt figure out how to compare the plots. which plot more stable?. I actually should know how to use b and d.
Any suggestion (including math link) is appreciated.
  1 个评论
Image Analyst
Image Analyst 2016-12-10
Show a plot of what your data look like. And, what function did you use to do the fitting? And was it the Curve Fitting Toolbox or the Statistics and Machine Learning Toolbox (please list it below in the products section).

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 Fit Postprocessing 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