Lesson19a Heteroskedasticity скачать в хорошем качестве

Lesson19a Heteroskedasticity

Boston University EE509 "Applied Environmental Statistics" Course: In this lecture we introduce graphical notation for Bayesian models and review the assumptions in a linear model assumption. We then dive into our first assumption (homoskedasticity = equal variance) and discuss how we can relax this assumption by explicitly modeling changes in variance (heteroskedasticity). We start with a simple linear model (including MLE R code and Bayesian JAGS code), show how model selection (AIC, DIC) can select variance models not just mean models, and then discuss how to generalize heteroskedastic models. http://people.bu.edu/dietze/Bayes2020...