From ac282fa2101fa2f9309b1912b4eb6fd561d48452 Mon Sep 17 00:00:00 2001 From: Nick Golding Date: Fri, 9 Aug 2019 10:49:25 +1000 Subject: [PATCH] remove link to Finnish horseshoe paper as projecteuclid is down --- docs/articles/example_models.html | 2 +- inst/examples/linear_finnish_horseshoe.Rmd | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/docs/articles/example_models.html b/docs/articles/example_models.html index 08d86b2f..60f05ea7 100644 --- a/docs/articles/example_models.html +++ b/docs/articles/example_models.html @@ -354,7 +354,7 @@

Regularized horseshoe prior

-

The regularized (‘Finnish’) horseshoe remedies a problem of the original horseshoe: large, unregularized values for the coefficients. This is especially problematic in scenarios where the parameters are only weakly identified by the data, as in logistic regression with perfectly seperable data.

+

The regularized (‘Finnish’) horseshoe (doi.org/10.1214/17-EJS1337SI) remedies a problem of the original horseshoe: large, unregularized values for the coefficients. This is especially problematic in scenarios where the parameters are only weakly identified by the data, as in logistic regression with perfectly seperable data.

regularized_horseshoe <- function (tau = 1,  c = 1, dim = NULL) {
   stopifnot(c > 0)
   lambda <- cauchy(0, 1, truncation = c(0, Inf), dim = dim)
diff --git a/inst/examples/linear_finnish_horseshoe.Rmd b/inst/examples/linear_finnish_horseshoe.Rmd
index 8fd33bd5..5c047f7c 100644
--- a/inst/examples/linear_finnish_horseshoe.Rmd
+++ b/inst/examples/linear_finnish_horseshoe.Rmd
@@ -1,6 +1,6 @@
 ### Regularized horseshoe prior
 
-The [regularized ('Finnish') horseshoe](https://projecteuclid.org/euclid.ejs/1513306866) remedies a problem of the original horseshoe: large, unregularized values for the coefficients. This is especially problematic in scenarios where the parameters are only weakly identified by the data, as in logistic regression with perfectly seperable data.
+The regularized ('Finnish') horseshoe (doi.org/10.1214/17-EJS1337SI) remedies a problem of the original horseshoe: large, unregularized values for the coefficients. This is especially problematic in scenarios where the parameters are only weakly identified by the data, as in logistic regression with perfectly seperable data.
 
 ```{r linear_finnish_horseshoe_greta}
 regularized_horseshoe <- function (tau = 1,  c = 1, dim = NULL) {