Questions about some results from Curve Fitting Tool
Mostrar comentarios más antiguos
After using the Curve Fitting Tool, I get the following output:
----------------------------------
General model:
f(x) = a*((x-3.4)^(1/2)*(1/(exp((x-3.4-2.92099 )/(8.6*10^(-5)*b))+1)))+c
Coefficients (with 95% confidence bounds):
a = -4243
b = 300
c = 1814 (1738, 1890)
Goodness of fit:
SSE: 6748
R-square: 0.9789
Adjusted R-square: 0.9759
RMSE: 21.95
-----------------------------------
My questions are:
- For coefficients a and b shown above, what is the meaning that the confidence bounds were not calculated?
- Can I trust these numbers?
1 comentario
Zain Mecklai
el 20 de En. de 2011
By the way, we support markup for making ordered lists using the # symbol at the start of a new paragraph
Like this:
# Note the blank line above
# This would be item 2
Respuestas (1)
neha
el 20 de Mzo. de 2012
0 votos
confidence bound can b changed...but by default.it takes d bound of 95%. and if your model get converged....plus u shud try to minimize RMSE value..it would be a better fit.
Categorías
Más información sobre Get Started with Curve Fitting Toolbox en Centro de ayuda y File Exchange.
Productos
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!