fit.WinCurse: Fit Winner's Curse Model

In zrmacc/WinCurse: Winners Curse Correction

Description

Fit Winner's Curse Model

Usage

fit.WinCurse(
  theta,
  se,
  pi = NULL,
  tau2 = NULL,
  eps = 1e-06,
  maxit = 100,
  report = FALSE
)

Arguments

theta	Parameter estimates.
se	Parameter standard errors.
pi	Initial value of pi, the proportion of null parameters.
tau2	Initial value of tau, the variance component.
eps	Tolerance for Newton-Raphson iterations.
maxit	Maximum number of NR iterations.
report	Report fitting progress?

Value

Object of class 'winCurse' containing:

• @Assignments: Maximum a posteriori component assignment and assignment entropy.

• @Estimates: Estimated model parameters.

• @Expectations: Posterior expected effect sizes.

• @Responsibilities: Probabilities the parameter came from the null and non-null components.

Examples

data(wc_data)
fit <- fit.WinCurse(
  theta = wc_data$theta,
  se = wc_data$se
)

zrmacc/WinCurse documentation built on April 2, 2021, 3:40 a.m.