just take relative URLs out of vignettes

DESCRIPTION

@@ -1,8 +1,8 @@
 Package: rstanarm
 Type: Package
 Title: Bayesian Applied Regression Modeling via Stan
-Version: 2.19.1
-Date: 2019-09-16
+Version: 2.19.2
+Date: 2019-10-01
 Encoding: UTF-8
 Authors@R: c(person("Jonah", "Gabry", email = "[email protected]", role = "aut"),
              person("Imad", "Ali", role = "ctb"),

vignettes/glmer.Rmd

@@ -22,9 +22,9 @@ params:
 This vignette explains how to use the `stan_lmer`, `stan_glmer`, `stan_nlmer`, 
 and `stan_gamm4` functions in the __rstanarm__ package to estimate linear and
 generalized (non-)linear models with parameters that may vary across groups.
-Before continuing, we recommend reading the [vignettes](glm.html) for the
-`stan_glm` function. The _Hierarchical Partial Pooling_ [vignette](pooling.html)
-also has examples of both `stan_glm` and `stan_glmer`.
+Before continuing, we recommend reading the vignettes (navigate up one level) for 
+the various ways to use the `stan_glm` function. The _Hierarchical Partial Pooling_ 
+vignette also has examples of both `stan_glm` and `stan_glmer`.
 
 # GLMs with group-specific terms
 

vignettes/lm.Rmd

@@ -330,7 +330,8 @@ distributions thoughtfully and take a short-cut by specifying one prior
 distribution that is taken to apply to all the regression coefficients as if
 they were independent of each other (and the intercept and error variance). 
 This short-cut is available in the `stan_glm` function and is described in more
-detail in other __rstanarm__ [vignette](glm.html) for Generalized Linear Models (GLMs).
+detail in other __rstanarm__ vignettes for Generalized Linear Models (GLMs),
+which can be found by navigating up one level.
 
 We are optimistic that this prior on the $R^2$ will greatly help in accomplishing 
 our goal for __rstanarm__ of making Bayesian estimation of regression models