Parametric GWAS mapping — param

Maps quantile-level GWAS results into effects on user-specified parametric system parameters via a weight matrix W.

Usage

param_gwas(
  stage1,
  transform = c("custom_W", "two_normal"),
  transform_args = list(),
  se_mode = c("diagonal", "plugin_cor", "dwls"),
  plugin_R = NULL
)

Arguments

stage1

Output list from quantile_gwas().

transform

One of:

"custom_W" – supply a weight matrix W in transform_args.
"two_normal" – construct weights for a two-component normal mixture.

transform_args

Arguments for the chosen transform.

se_mode

Standard error mode:

"diagonal" – assume independence across taus (fast).
"plugin_cor" – plugin correlation with near-PD repair.
"dwls" – diagonal weighted least squares (Q test calibrated).

plugin_R

Optional T x T correlation matrix for "plugin_cor".

Value

A list with the following elements:

W: Weight matrix used in the parametric mapping.
A: Mapping matrix (tau-by-parameter).
params: Estimated parameter values.
SE_params: Delta-method standard errors of parameter estimates.
Q: Vector of Q statistics per SNP (if computed).
df: Degrees of freedom for Q statistic.