MODEL ADITIF TERGENERALISASI
Modeling relationship between respond variable and predictor is not always following linearity and normality assumption. Hastie and Tibshirani (1986) adapted additive models to generalized linear models that is called by generalized additive models. Generalized additive models is modeling technique...
محفوظ في:
| المؤلفون الرئيسيون: | , |
|---|---|
| التنسيق: | Theses and Dissertations NonPeerReviewed |
| منشور في: |
[Yogyakarta] : Universitas Gadjah Mada
2013
|
| الموضوعات: | |
| الوصول للمادة أونلاين: | https://repository.ugm.ac.id/122894/ http://etd.ugm.ac.id/index.php?mod=penelitian_detail&sub=PenelitianDetail&act=view&typ=html&buku_id=63003 |
| الوسوم: |
إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
|
| الملخص: | Modeling relationship between respond variable and predictor is not always
following linearity and normality assumption. Hastie and Tibshirani (1986) adapted
additive models to generalized linear models that is called by generalized additive
models. Generalized additive models is modeling technique that appropriates for
overcomes nonlinearity in the relationship between respond variable and predictor,
and does not limited respond variable to normal distribution but other distributions
in the exponential family allowing to use in this model.
Generalized additive models replace linear component on the generalized
linear models with sum of functions which is estimated using local scoring algorithm.
Additive component in generalized additive models is sum of univariat function of
every predictors, so we can see contribution of each predictor to respond. Illustration
is given in case study using R software. |
|---|