Hello. I am doing population pharmacokinetic analysis using Simbiology. There are 4 types of error model available in Simbiology. Constant (y = f+a*e), exponential (y = exp(a*e)), combined (y = f + (a+b*f)*e) and proporttional (y = f+b*f*e).
Suppose i want to create new error model say
y = f * (1+a)+b
How will I be able to achieve it.
Regards.

 Respuesta aceptada

Jeremy Huard
Jeremy Huard el 1 de Jul. de 2019

0 votos

Hi Praveen,
the error model you suggested looks like a combined error model. But from your equation I don't see whether you expect one or two normally distributed random variable(s). How does the error term look like?
If it is (a+b*|f|)*e, then it's exactly SimBiology's combined error model.
If it is (a*e1 + b*|f|*e2), then unfortunately it can't be implemented in SimBiology as SimBiology does not support custom error models as of release R2019a.
Best,
Jérémy

1 comentario

Praveen Kumar M
Praveen Kumar M el 3 de Jul. de 2019
Thanks Jeremy for the clarification. I suppose that I can proceed forward with combined error model as suggested.

Iniciar sesión para comentar.

Más respuestas (0)

Comunidades de usuarios

Más respuestas en  SimBiology Community

Categorías

Etiquetas

Preguntada:

el 1 de Jul. de 2019

Comentada:

el 3 de Jul. de 2019

Community Treasure Hunt

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

Start Hunting!

Translated by