Mixed Effects Kernel Bernoulli Regression

• 발행기관: 한국자료분석학회
• 저자명: 김대학(Daehak Kim)
• 간행물 정보: 『Journal of The Korean Data Analysis Society (JKDAS)』Vol.10 No.6, 3067~3076쪽, 전체 10쪽
• 주제분류: 자연과학 > 통계학
• 파일형태: PDF
• 발행일자: 2008.12.30

영문 초록

For the mixed effects nonlinear Bernoulli regression, an estimating procedure of the success probability is introduced, which is based on the penalized negative log-likelihood including random effects term. The proposed procedure provides the estimates of the success probability of the response variables, where the canonical parameter is es expressed as the sum of the nonlinear function of input vectors and the random effects. The generalized cross validation function is introduced to choose optimal hyper-parameters in the procedure. Numerical studies are performed through the simulated data set and the real data set to indicate the performance of the proposed estimating procedure.

목차

1. Introduction

2. Nonlinear Bernoulli Regression

3. Mixed Effect Kernel Bernoulli Regression

4. Model Selection

5. Numerical Studies

6. Conclusions

References

키워드

Generalized linear model Bernoulli distribution Penalized negative log-likelihood Mixed effects model Generalized cross validation function