Transfer function estimation tfestOptions

3 次查看（过去 30 天）
Hussein
Hussein 2023-11-25
评论： Hussein 2023-12-12
Hello,
I am using tfest to estimate transfer function in frequency domain and time domain. The tfestOptions that I am using are the following:
opt = tfestOptions('InitialCondition', 'estimate', ...
'SearchMethod', 'lm', ...
'Display', 'on', ...
'EstimateCovariance', true)
I was wondering how can I know the cost function that is used according to the options used above? Also, how can I use different cost functions if i want?
Thank you for your help.
  5 个评论
显示 3更早的评论
Mathieu NOE
Mathieu NOE 2023-11-27
well I don't know if you want to share your data and code and maybe someone will find something interesting to share with you (?)
Hussein
Hussein 2023-11-27
Unfortunately i can't share data, but thanks anyways

请先登录，再进行评论。

请先登录，再回答此问题。

回答（1 个）

Rishi
Rishi 2023-12-7
Hi Hussein,
I understand from your query that you want to know what cost function you are using, and how to use a different cost function.
The cost function can be configured according to your needs by tinkering with the following estimation options:
  • Focus
  • WeightingFilter
  • EnforceStability
  • OutputWeight
  • ErrorThreshold
  • Regularization
These options can be modified in ‘tfestOptions’, except ‘Focus’. You can learn more about how each of these options affects the cost function from the below documentation:
https://www.mathworks.com/help/releases/R2023a/ident/ug/model-quality-metrics.html#:~:text=and%20OutputWeight.-,Options%20to%20Configure%20the%20Loss%20Function,-You%20can%20configure
Also, you can learn more about ‘tfestOptions’ from the below documentation:
https://www.mathworks.com/help/releases/R2023a/ident/ref/tfestoptions.html
Hope this helps.
  1 个评论
Hussein
Hussein 2023-12-12
Thank your for you respond.
I have already seen this document. In the documentation it shows that a generic cost function is used. My aim is to change or modify the cost function

请先登录，再进行评论。

类别

在 Help Center 和 File Exchange 中查找有关 Transfer Function Models 的更多信息

标签

产品


版本

R2023a

Community Treasure Hunt

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

Start Hunting!

Translated by