0.2.1:

	Bug correction: handles special case where one predictor is identical with target variable.

UPDATES

@@ -1,3 +1,5 @@
+0.2.1:
+	Bug correction: handle special cases where one predictor is identical with target variable.
 0.2.0:
 	Added an additional null distribution: spherical, besides the original normal distribution. This can be set in the new H0 parameter. For details, see reference.
 	Corrected normalize=FALSE cases.

lassopv/DESCRIPTION

@@ -1,14 +1,14 @@
 Package: lassopv
 Type: Package
 Title: Nonparametric P-Value Estimation for Predictors in Lasso
-Version: 0.2.0
-Date: 2018-02-22
+Version: 0.2.1
+Date: 2019-12-03
 Author: Lingfei Wang <[email protected]>
 Maintainer: Lingfei Wang <[email protected]>
 Description: Estimate the p-values for predictors x against target variable y in lasso regression, using the regularization strength when each predictor enters the active set of regularization path for the first time as the statistic. This is based on the assumption that predictors (of the same variance) that (first) become active earlier tend to be more significant. Three null distributions are supported: normal and spherical, which are computed separately for each predictor and analytically under approximation, which aims at efficiency and accuracy for small p-values.
 URL: https://github.com/lingfeiwang/lassopv
 License: GPL-3
-Copyright: Copyright 2016-2018 Lingfei Wang
+Copyright: Copyright 2016-2019 Lingfei Wang
 Depends: R (>= 2.10)
 Imports: lars, stats
 NeedsCompilation: no

lassopv/R/lassopv.R

@@ -1,4 +1,4 @@
-# Copyright 2016-2018 Lingfei Wang
+# Copyright 2016-2019 Lingfei Wang
 # 
 # This file is part of lassopv.
 # 
@@ -54,7 +54,12 @@ lassopv<-function (x,y,normalize=TRUE,H0=c("spherical","normal"),log.p=FALSE,max
 		#lambda values
 		lambda=a$lambda[1:nstep,drop=F]/ds
 		#Residue
-		yres=y-predict.lars(a,x,s=a$lambda[1:nstep,drop=F],mode='lambda')$fit
+		ypredict=predict.lars(a,x,s=a$lambda[1:nstep,drop=F],mode='lambda')$fit
+		if(is.null(dim(ypredict)))
+		{
+			ypredict=matrix(ypredict,ncol=1)
+		}
+		yres=y-ypredict
 		#Residue variance
 		pv=t(t(yres)-colMeans(yres))
 		pv=colMeans(pv**2)
@@ -97,12 +102,14 @@ lassopv<-function (x,y,normalize=TRUE,H0=c("spherical","normal"),log.p=FALSE,max
 	if(log.p)
 	{
 		ans[!isdone]=0
+		ans[is.nan(ans)]=-Inf
 		#Account for numerical precision limitations
 		ans[ans>0]=0
 	}
 	else
 	{
 		ans[!isdone]=1
+		ans[is.nan(ans)]=0
 		#Account for numerical precision limitations
 		ans[ans>1]=1
 	}