data_lnln {hbsaems} | R Documentation |
Simulation Data for Lognormal-Lognormal Model
Description
This dataset is a simulated data created for demonstrating the implementation of Hierarchical Bayesian Small Area Estimation (HB SAE) using a lognormal-lognormal model. It includes area-level covariates, random effects, direct estimates, and spatial components, for testing SAE models with lognormal assumptions and spatial correlation.
Usage
data_lnln
Format
A data frame with 100 rows and 13 variables:
- group
Area ID (1–100) for random effects formula specifying the grouping structure in the data.
- x1, x2, x3
Auxiliary area-level covariates
- u_true
True unstructured area-level random effect on the log scale.
- teta_true
True linear predictor on the log scale (meanlog for lognormal distribution).
- mu_orig_true
True mean on the original scale, calculated from
eta_true
andsigma_e
.- n
Sample size per area.
- y_obs
Simulated observed mean per area, generated from a lognormal distribution.
- lambda_dir
Direct estimator of the mean per area (same as
y_obs
).- y_log_obs
Log-transformed direct estimator.
- psi_i
Approximate sampling variance of
y_obs
.- sre
An optional grouping factor mapping observations to spatial locations.
Source
Simulated data based on a Lognormal–Lognormal model.