Repglmre2: Loop through all locations and run GLMM for each

In DHSr: Create Large Scale Repeated Regression Summary Statistics Dataset and Visualization Seamlessly

Loop through all locations and run GLMM for each

Description

This function runs a mixed-effects generalized linear model (GLMM) for each location within a dataset.

Usage

Repglmre2(data, formula, location_var, random_effect_var, family)

Arguments

data	The dataset to be analyzed.
formula	The formula for the regression model.
location_var	The variable indicating different locations (e.g., 'REGCODE').
random_effect_var	The variable to be used as a random effect (e.g., 'hhid').
family	The family to be used for GLM (e.g., 'binomial' for logistic regression, 'poisson' for Poisson regression).

Value

A dataframe containing the results

Examples

set.seed(123)

# Create dummy data
  library(dplyr)
dummy_data <- data.frame(
  years_education = rnorm(100, 12, 3),    # Represents years of education
  gender_female = rbinom(100, 1, 0.5),    # 1 = Female, 0 = Male
  household_wealth = sample(1:5, 100, replace = TRUE),  # Wealth index from 1 to 5
  district_code = sample(1:10, 100, replace = TRUE)     # Represents district codes
) %>% arrange(district_code)

# Create HHid (Household ID), grouping every 3-4 records, and convert to character
dummy_data$HHid <- as.character(rep(1:20, each = 5, length.out = nrow(dummy_data)))

# Create a binary outcome variable for years of education
dummy_data$education_binary <- ifelse(dummy_data$years_education > 11, 1, 0)

# Define a logistic regression formula
formula <- education_binary ~ gender_female + household_wealth:gender_female

location_var <- "district_code"
random_effect_var <- "HHid"

# Run the logistic mixed-effects model across all locations (districts)
results <- DHSr::Repglmre2(data = dummy_data, formula = formula,
                           location_var = location_var, random_effect_var = random_effect_var,
                           family = binomial())

# Print the results
print(head(results))

DHSr documentation built on April 4, 2025, 12:18 a.m.