bug

Author: Jelena Markovic

selectiveInference/R/funs.randomized.R

@@ -258,8 +258,7 @@ conditional_density = function(noise_scale, lasso_soln) {
   return(lasso_soln)
 }
 
-randomized_inference = function(X, y, sigma, lam, noise_scale, ridge_term, 
-                                condition_subgrad=FALSE, level=0.9){
+randomized_inference = function(X, y, sigma, lam, noise_scale, ridge_term, level=0.9){
 
   n = nrow(X)
   p = ncol(X)
@@ -268,9 +267,7 @@ randomized_inference = function(X, y, sigma, lam, noise_scale, ridge_term,
   inactive_set = lasso_soln$inactive_set
   nactive = length(active_set)
 
-  if (condition_subgrad==TRUE){
-    lasso_soln=conditional_density(noise_scale,lasso_soln)
-  } 
+  lasso_soln=conditional_density(noise_scale,lasso_soln)
 
   dim = length(lasso_soln$observed_opt_state)
   print(paste("chain dim", dim))
@@ -296,7 +293,7 @@ randomized_inference = function(X, y, sigma, lam, noise_scale, ridge_term,
                                                   target_cov[i,i], 
                                                   cov_target_internal[,i],
                                                   internal_transform)
-    target_sample = rnorm(nrow(opt_samples)) * sqrt(target_cov[i,i])
+    target_sample = rnorm(nrow(as.matrix(opt_samples))) * sqrt(target_cov[i,i])
 
     pivot = function(candidate){
       weights = selectiveInference:::importance_weight(noise_scale,

tests/randomized/test_instances.R

@@ -25,9 +25,9 @@ gaussian_instance = function(n, p, s, sigma=1, rho=0, signal=6, X=NA,
 }
 
 
-collect_results = function(n,p,s, nsim=100, level=0.9){
+collect_results = function(n,p,s, nsim=1, level=0.9){
   rho=0.3
-  lam=1.
+  lam=2.
   sigma=1
   sample_pvalues = c()
   sample_coverage = c()
@@ -38,7 +38,7 @@ collect_results = function(n,p,s, nsim=100, level=0.9){
     beta=data$beta
     ridge_term=sd(y)/sqrt(n)
     noise_scale = sd(y)/2
-    result = selectiveInference:::randomized_inference(X,y,sigma,lam,noise_scale,ridge_term, TRUE, level)
+    result = selectiveInference:::randomized_inference(X,y,sigma,lam,noise_scale,ridge_term, level)
     true_beta = beta[result$active_set]
     coverage = rep(0, nrow(result$ci))
     for (i in 1:nrow(result$ci)){