A short explanation of inla.stack() for hierarchical modelling | Classical and Berkson measurement error and missingness | A model for missing data, missing not at random

First example: A logistic regression model with repeated measurements | Second example: Heteroscedastic measurement error and interaction between error variable and error free variable

Simple example with missingness and two types of measurement error | Generating the data | Fitting the model | Missing data only | Model without imputation | Model with imputation | Random effect in the main model | Interaction effect with error variable | Logistic regression with classical error and missing data | Poisson regression with classical error and missing data

Defining the model: formula + structured data stacks | Structure for a classical measurement error model | $$\underbrace | Structure for Berkson and classical measurement error model | $$\underbrace | Accessing the stacks from the model object | $$\underbrace