How to find the erro for fitted parameters?

15 vues (au cours des 30 derniers jours)
John
John le 15 Nov 2017
Commenté : Star Strider le 15 Nov 2017
When using 'lsqnonlin', 'lsqcurvefit', 'fitnlm', 'nlinfit', etc.. It produces optimized parameters. How to find the error of these parameters to report results like a+/- da; b +/- db ( da and db here)? Thanks.

Réponse acceptée

Star Strider
Star Strider le 15 Nov 2017
To get parameter confidence limits on a nonlinear regression, use the nlparci (link) function.
To get confidence intervals on the fit (prediction), use the nlpredci (link) function.
  2 commentaires
John
John le 15 Nov 2017
Thanks Star!
Star Strider
Star Strider le 15 Nov 2017
My pleasure!

Connectez-vous pour commenter.

Plus de réponses (0)

Catégories

En savoir plus sur Get Started with Curve Fitting Toolbox dans Help Center et File Exchange

Produits

Community Treasure Hunt

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

Start Hunting!

Translated by