\(\bar y + b(\bar X - \bar x)\). Unlike the ratio estimator it
fits an INTERCEPT rather than forcing the line through the origin,
and its variance is \((1 - \rho^2)\) times the simple-mean
variance, so the gain is exactly the squared correlation. Mirrors
morie.fn.regest.
Usage
morie_regression_estimator(y, x, weights = NULL, X_mean = NULL)
Arguments
- y, x
study and auxiliary variables.
- weights
design weights.
- X_mean
known population mean of x; required.
Value
list: mean, slope, intercept, correlation,
variance_ratio_to_simple_mean, passes_through_origin, n, method.
Examples
set.seed(1)
x <- runif(60, 5, 15)
morie_regression_estimator(100 + 2 * x + rnorm(60), x, X_mean = 10)$slope
#> [1] 1.989036
