Scholarlink

In this chapter, the models are developed for correlated outcomes where the association parameters are of interest. The models are based on marginal-conditional approachMarginal-conditional approach and generalized linear models are developed for discrete and continuous random variables. Islam and Chowdhury, PLoS ONE 12, (2017a), Islam and Chowdhury, Analysis of repeated measures data, Springer (2017b) provided introduction to bivariate models for different types of data. The conditional models are presented along with bivariate models using multiplication where joint probability distribution is shown by multiplying the conditional probability with marginal probability. The exponential forms, generalized linear model representations, over-dispersion problems and remedial measures, measures of goodness of fit, devianceDeviance, and testing of hypotheses are discussed with emphasis to make the users understand the concepts without difficulty.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Extension of GLM for Bivariate Data

  • 章节

  • 发表日期：

    2025年

查看全文
  • M. Ataharul Islam,
  • Soma Chowdhury Biswas

摘要

In this chapter, the models are developed for correlated outcomes where the association parameters are of interest. The models are based on marginal-conditional approachMarginal-conditional approach and generalized linear models are developed for discrete and continuous random variables. Islam and Chowdhury, PLoS ONE 12, (2017a), Islam and Chowdhury, Analysis of repeated measures data, Springer (2017b) provided introduction to bivariate models for different types of data. The conditional models are presented along with bivariate models using multiplication where joint probability distribution is shown by multiplying the conditional probability with marginal probability. The exponential forms, generalized linear model representations, over-dispersion problems and remedial measures, measures of goodness of fit, devianceDeviance, and testing of hypotheses are discussed with emphasis to make the users understand the concepts without difficulty.