Finding AIC for an exponential model

11 次查看（过去 30 天）
Alice Lam
Alice Lam 2021-7-30
回答： Phil Steindel 2021-7-30
Is there any way to find the AIC for an exponential model? I know fitlm gives me the AIC for a linear regression, but I can't seem to calculate the AIC for an exponential fit as described by the code at the bottom. I have tried expfit, which returned a single parameter, and the aic() function, but I can't figure out how to turn this into a state-space model that would work with the aic function.
exp_fit = fittype(@(a,b,c,d,x) a+b*exp(c*x+d));
exp_fit_options = fitoptions(exp_fit);
exp_fit_options.Upper = [Inf, Inf, Inf];
exp_fit_options.Lower = [-Inf, 0, 0];
linear_model = fitlm(x,y)
exp_model = fit(x, y, exp_fit)

请先登录，再回答此问题。

回答（1 个）

Phil Steindel
Phil Steindel 2021-7-30
Hi Alice,
Assuming you have access to Machine Learning Toolbox (as you've used fitlm and expfit), you can try fitnlm. As with fitlm, the AIC is in the ModelCriterion property of the output. Note that you will also have to specify an initial guess for the parameters, and you won't be able to specify upper and lower bounds for them as you have done here.

类别

在 Help Center 和 File Exchange 中查找有关 Fit Postprocessing 的更多信息

标签

产品


版本

R2021a

Community Treasure Hunt

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

Start Hunting!

Translated by