BLUP of a random intercept

uhat_j = (s2u / (s2u + s2e/n_j)) * mean_j(y - X beta).

Usage

Blupint(y, group, s2u, s2e, X = NULL, beta = NULL)

Arguments

y

Response.

group

Group label per observation.

s2u

Between-group variance, non-negative.

s2e

Within-group variance, strictly positive.

X

Fixed-effect design, or NULL.

beta

Fixed-effect coefficients, required when X is given.

Value

List with u, shrink, nj, groupmean, levels, vpc, J, n.

References

Henderson (1975), Biometrics 31(2), 423-447; Robinson (1991), Statistical Science 6(1), 15-32. Standard published form; neither article is in the local corpus and neither was read.

Examples

set.seed(1)
Blupint(y = rnorm(30), group = rep(1:5, each = 6), s2u = 1, s2e = 1)
#> $u
#> [1] -0.02487413  0.48539588 -0.11832817  0.17163564 -0.16043705
#> 
#> $shrink
#> [1] 0.8571429 0.8571429 0.8571429 0.8571429 0.8571429
#> 
#> $nj
#> [1] 6 6 6 6 6
#> 
#> $groupmean
#> [1] -0.02901982  0.56629519 -0.13804954  0.20024158 -0.18717656
#> 
#> $levels
#> [1] "1" "2" "3" "4" "5"
#> 
#> $vpc
#> [1] 0.5
#> 
#> $J
#> [1] 5
#> 
#> $n
#> [1] 30
#> 
#> $method
#> [1] "BLUP of a random intercept (Henderson 1975; Robinson 1991)"
#> 
#> attr(,"class")
#> [1] "morie_rich_result" "list"