Marker variance-component estimation

Reports sigma_m^2 via the VanRaden split sigma_g^2 / (2 sum p_j q_j) alongside the naive sigma_g^2 / p form. sigma_g^2 is obtained from a quick GBLUP fit.

Usage

morie_marker_variance(x, y, markers)

Arguments

x

Fixed-effect design (optional).

y

Numeric response.

markers

(n x m) genotype matrix coded 0/1/2.

Value

list(estimate, sigma_g2, sigma_e2, h2, sigma_m2_vanraden, sigma_m2_naive, sum_2pq, p_freq, n, p, method).

References

VanRaden (2008); Montesinos Lopez Ch 3.

Examples

set.seed(1)
morie_marker_variance(
  x = rnorm(50), y = rnorm(50),
  markers = matrix(sample(0:2, 200, TRUE), 50, 4)
)
#> $estimate
#> [1] 0.0386784
#> 
#> $sigma_g2
#> [1] 0.07533005
#> 
#> $sigma_e2
#> [1] 0.5851633
#> 
#> $h2
#> [1] 0.1140512
#> 
#> $sigma_m2_vanraden
#> [1] 0.0386784
#> 
#> $sigma_m2_naive
#> [1] 0.01883251
#> 
#> $sum_2pq
#> [1] 1.9476
#> 
#> $p_freq
#> [1] 0.49 0.57 0.36 0.46
#> 
#> $n
#> [1] 50
#> 
#> $p
#> [1] 4
#> 
#> $method
#> [1] "VanRaden + naive marker-variance split"
#>