LogLike: Approximate log likelihood of an experiment.

In gpapadog/LERCA: Local Exposure-Response Confounding Adjustment

Description

Calculating the log likelihood based on the BIC approximation.

Usage

LogLike(D, curr_exper_alphas, curr_coefsY = NULL, X_s_cut, cov_cols)

Arguments

D	The data set of the current experiment includes covariates, exposure as 'X' and outcome as 'Y'.
curr_exper_alphas	Matrix. Dimensions correspond to exposure/outcome model and potential confounders. Entries 0/1 represent exlusion/inclusion of the covariate in each model.
curr_coefsY	Numeric of length two. Intercept and slope of the outcome model. If left NULL, the likelihood is calculated integrating out intercept and slope along with remaining coefficients.
X_s_cut	Numeric. The point in the experiment configuration that corresponds to the beginning of the current experiment.
cov_cols	The indices of the columns in D that correspond to the potential confounders.

gpapadog/LERCA documentation built on June 4, 2019, 11:40 a.m.