I working on the development a statistical model to reproduce the performance of an EC reactor. I need to assess the influence of each single parameter, to do so, I used the Beta coefficient. Dose anyone know another reliable method please?
Dear Doctor,
You can use the cosine amplitude method (CAM) to evaluate the relative effect of each input factor in the network system on the output factors as suggested by Yang and Zhan
Yang Y, Zhang Q (1997) A hierarchical analysis for rock engineering
using artificial neural networks. Rock Mech Rock Eng
30:207–222. doi:10.1007/BF01045717
I hope it will be useful for you
Kind regards,
Salah
If this is a regression model, you can use the confidence interval of the beta coefficient. Also the R-squared change in the model that happens if the coefficient is removed.
Dear Dr. Jerry
Many thanks for response, yes it is a regression model.
A response surface methodology can be used to find a polynomial regression model with careful experiments design selection that assess and check each parameter of process.
Dear Doctor,
You can use the cosine amplitude method (CAM) to evaluate the relative effect of each input factor in the network system on the output factors as suggested by Yang and Zhan
Yang Y, Zhang Q (1997) A hierarchical analysis for rock engineering
using artificial neural networks. Rock Mech Rock Eng
30:207–222. doi:10.1007/BF01045717
I hope it will be useful for you
Kind regards,
Salah
Dear Khalid
You can use a t-test for testing the significance of beta coefficient. You can also use confidence interval approach, if it doesn't contain zero. Coefficient of determination R^2 adjusted and Mean Square Errors with and without including the particular factor in regression model can also be used.
Regards
You can use a t-test for testing the significance of beta coefficient. You can also use the confidence interval approach, if it doesn't contain zero.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Statistical Modeling