MODEL ADITIF CAMPURAN TERGENERALISASI
The generalized additive models (GAM) is an extension of the usual linear regression by generalizing linear functions into an additive function so that this model can be used even though the relationship of the response variable and linear predictor variables is not linear. And the response variable...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Theses and Dissertations NonPeerReviewed |
| Published: |
[Yogyakarta] : Universitas Gadjah Mada
2014
|
| Subjects: | |
| Online Access: | https://repository.ugm.ac.id/131549/ http://etd.ugm.ac.id/index.php?mod=penelitian_detail&sub=PenelitianDetail&act=view&typ=html&buku_id=72045 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The generalized additive models (GAM) is an extension of the usual linear regression by generalizing linear functions into an additive function so that this model can be used even though the relationship of the response variable and linear predictor variables is not linear. And the response variable exponential family. But GAM can not be used if there are two effect influencing the model, that is fixed effect and random effect. The generalized additive mixed models (GAMM) is expected to be more efficient in identifying the effect of the distribution of the random component that is able to explain precisely the effect of a random component in a model. The use of generalized additive mixed models for quantitative variables with the estimation of data smoothing using smoothing spline functions, and parameter estimation using Maximum Likelihood Estimation (MLE) can not be solved analytically, so the estimator is computed by maximizing the log-likelihood function numerically using the Newton-Raphson method with using the expected second derivative of the log-likelihood function called fisher scoring techniques. |
|---|