Data
====

radon.data.R
Source: radon measurements coordinated by the Environmental Protection Agency
in a random sample of more than 80,000 houses throughout the country.
The dataset is restricted to the 85 counties in Minnesota.
- N          : number of observations
- J          : number of counties
- y          : logarithm of the radon measurement in house i, i = 1,...,N
- x          : the floor of the measurement
               0: basement
               1: first floor
- county     : county id, 1,...,J
- county_name: county name
- u          : measurement of soil uranium in county j, j = 1,...,J

Models
======

1. radon.1.stan: varying intercept multilevel model

2. radon.pooling.stan: complete pooling model

3. radon.nopooling.stan: no pooling model

4. radon.2.stan: multilevel model with group-level predictors

5. radon.2a.stan: predicting a new unit in an existing group

6. radon.2b.stan: predicting a new unit in a new group

7. radon.3.stan: specifying the values of sigma.y and sigma.a