rebuild from rstantools

.Rbuildignore

@@ -1,6 +1,7 @@
+^rcppExports.cpp$
+^stanExports_*
+
 ^\.travis\.yml$
-rccpExports.cpp
-stanExports_*
 ^data-raw$
 .Rhistory
 README.Rmd
@@ -18,5 +19,4 @@ tests/testthat/test-rstan-conformity-results
 NEWS.md
 index.html
 cran-comments.md
-
 ^CRAN-RELEASE$

.gitignore

@@ -1,9 +1,14 @@
-src/geostan.so
-src/geostan.dll
+^rcppExports.cpp$
+^stanExports_*
 !man/geostan-internal.Rd
 !man/figures/logo.png
 man/*.Rd
 .Rhistory
-*~
-inst/doc
-cran-comments.md
+*~$
+cran-comments.md
+
+src/geostan.so
+src/geostan.dll
+man/*.Rd
+.Rhistory
+cran-comments.md

DESCRIPTION

@@ -15,10 +15,12 @@ Description: For Bayesian inference with spatial data, provides exploratory anal
 License: GPL (>= 3)
 Encoding: UTF-8
 LazyData: true
+Roxygen: list(markdown = TRUE)
+RoxygenNote: 7.1.1
 Biarch: true
 Depends: 
     R (>= 3.4.0)
-Imports:
+Imports: 
     spdep (>= 1.1-8),
     sf,
     ggplot2 (>= 3.0.0),
@@ -33,20 +35,14 @@ Imports:
     utils,
     Matrix (>= 1.3),
     Rcpp (>= 0.12.0),
-    RcppParallel,	    
+    RcppParallel (>= 5.0.1),
     rstan (>= 2.18.1),
-    rstantools (>= 2.0.0)
+    rstantools (>= 2.1.1)
 LinkingTo: 
     BH (>= 1.66.0),
     Rcpp (>= 0.12.0),
     RcppEigen (>= 0.3.3.3.0),
+    RcppParallel (>= 5.0.1),
     rstan (>= 2.18.1),
-    StanHeaders (>= 2.18.0),
-    RcppParallel     
-Suggests:
-    testthat,
-    knitr,
-    rmarkdown
+    StanHeaders (>= 2.18.0)
 SystemRequirements: GNU make
-RoxygenNote: 7.1.1
-VignetteBuilder: knitr

NEWS.md

@@ -4,7 +4,7 @@
 
 geostan now supports Poisson models with censored count data, a common problem in public health research where small area disease and mortality counts are censored below a threshold value. Model for censored outcome data can now be implemented using the `censor_point` argument found in all of the model fitting functions (stan_glm, stan_car, stan_esf, stan_icar).
 
-## measurement error models improved
+## Measurement error models improved
 
 The measurement error models have been updated in three important respects:
 

R/convenience-functions.R

@@ -328,7 +328,8 @@ lisa <- function(x, w, type = TRUE) {
 #' @param shape An object of class \code{sf} or another spatial object coercible to \code{sf} with \code{sf::st_as_sf} such as \code{SpatialPolygonsDataFrame}.
 #' @param name The name to use on the plot labels; default to "y" or, if \code{y} is a \code{geostan_fit} object, to "Residuals".
 #' @param plot If \code{FALSE}, return a list of \code{gg} plots.
-#' @param mc Character string indicating how to plot the residual Moran coefficient: if `mc = "scatter"`, then \code{\link[geostan]{moran_plot}} will be used with the marginal residuals; if `mc = "hist"`, then a histogram of Moran coefficient values will be returned, where each plotted value represents the degree of residual autocorrelation in a draw from the join posterior distribution of model parameters.
+#' 
+#' @param mc_style Character string indicating how to plot the residual Moran coefficient (only used if `y` is a fitted model): if `mc = "scatter"`, then \code{\link[geostan]{moran_plot}} will be used with the marginal residuals; if `mc = "hist"`, then a histogram of Moran coefficient values will be returned, where each plotted value represents the degree of residual autocorrelation in a draw from the join posterior distribution of model parameters.
 #' 
 #' @param style Style of connectivity matrix; if `w` is not provided, `style` is passed to \code{\link[geostan]{shape2mat}} and defaults to "W" for row-standardized.
 #' @param w An optional spatial connectivity matrix; if not provided, one will be created using \code{\link[geostan]{shape2mat}}.
@@ -368,7 +369,7 @@ sp_diag <- function(y,
                    shape,
                    name = "y",
                    plot = TRUE,
-                   mc = c("scatter", "hist"),
+                   mc_style = c("scatter", "hist"),
                    style = c("W", "B"),
                    w = shape2mat(shape, match.arg(style)),
                    binwidth = function(x) 0.5 * sd(x),
@@ -393,14 +394,14 @@ sp_diag.geostan_fit <- function(y,
                             shape,
                             name = "Residual",
                             plot = TRUE,
-                            mc = c("scatter", "hist"),                            
+                            mc_style = c("scatter", "hist"),                            
                             style = c("W", "B"),
                             w = shape2mat(shape, match.arg(style)),
                             binwidth = function(x) 0.5 * stats::sd(x),
                             rates = TRUE,
                             size = 0.15,
                             ...) {
-    mc <- match.arg(mc)
+    mc_style <- match.arg(mc_style, c("scatter", "hist"))
     if (!inherits(shape, "sf")) shape <- sf::st_as_sf(shape)
     outcome <- y$data[,1] 
     fits <- fitted(y, summary = TRUE, rates = rates)
@@ -432,13 +433,9 @@ sp_diag.geostan_fit <- function(y,
                              label = signs::signs) +
         theme_void()
     # residual autocorrelation
-
     R <- residuals(y, summary = FALSE)    
     R.mc <- apply(R, 1, mc, w = w)
-    if (length(unique(R.mc)) == 1) {
-        g.mc <- moran_plot(R[1,], w, xlab = name)
-    }
-    if (mc == "scatter") {
+    if (mc_style == "scatter") {
         g.mc <- moran_plot(marginal_residual, w, xlab = name)
     } else {
         R.mc.mu <- mean(R.mc)
@@ -453,6 +450,9 @@ sp_diag.geostan_fit <- function(y,
                 x = "Residual MC",
                 subtitle = paste0("MC (mean) = ", round(R.mc.mu, 2)))
     }
+    if (length(unique(R.mc)) == 1) {
+        g.mc <- moran_plot(R[1,], w, xlab = name)
+    }    
     if (plot) {
         return( gridExtra::grid.arrange(ovf, g.mc, map.y, ncol = 3) )
     } else {        
@@ -512,7 +512,7 @@ sp_diag.numeric <- function(y,
 #' @param probs Lower and upper quantiles of the credible interval to plot. 
 #' @param plot If \code{FALSE}, return a list of \code{ggplot}s and a \code{data.frame} with the raw data values alongside a posterior summary of the modeled variable.
 #'
-#' @param mc Character string indicating how to plot the Moran coefficient for the delta values: if `mc = "scatter"`, then \code{\link[geostan]{moran_plot}} will be used with the marginal residuals; if `mc = "hist"`, then a histogram of Moran coefficient values will be returned, where each plotted value represents the degree of residual autocorrelation in a draw from the join posterior distribution of delta values.
+#' @param mc_style Character string indicating how to plot the Moran coefficient for the delta values: if `mc = "scatter"`, then \code{\link[geostan]{moran_plot}} will be used with the marginal residuals; if `mc = "hist"`, then a histogram of Moran coefficient values will be returned, where each plotted value represents the degree of residual autocorrelation in a draw from the join posterior distribution of delta values.
 #' 
 #' @param size Size of points and lines, passed to \code{geom_pointrange}.
 #' @param index Integer value; use this if you wish to identify observations with the largest `n=index` absolute Delta values; data on the top `n=index` observations ordered by absolute Delta value will be printed to the console and the plots will be labeled with the indices of the identified observations.
@@ -570,7 +570,7 @@ me_diag <- function(fit,
                     shape,
                     probs = c(0.025, 0.975),
                     plot = TRUE,
-                    mc = c("scatter", "hist"),
+                    mc_style = c("scatter", "hist"),
                     size = 0.25,
                     index = 0,
                     style = c("W", "B"),
@@ -580,7 +580,7 @@ me_diag <- function(fit,
     stopifnot(length(varname) == 1)    
     if (!varname %in% colnames(fit$data)) stop("varname is not found in colnames(fit$data). Provide the name of the variable as it appears in the model formula")
     if (!inherits(shape, "sf")) shape <- sf::st_as_sf(shape)
-    mc <- match.arg(mc)
+    mc_style <- match.arg(mc_style, c("scatter", "hist"))
     x.raw <- as.numeric(fit$data[,varname])
     probs = sort(probs)
     width = paste0(100 * (probs[2] - probs[1]), "%")     
@@ -624,11 +624,11 @@ me_diag <- function(fit,
         theme_classic()
     delta.mat <- t(apply(x.samples, 1, .resid, y = x.raw))
     df$Delta <- apply(delta.mat, 2, mean)
-    if (mc == "scatter") {
+    D.mc <- apply(delta.mat, 1, mc, w = w)
+    D.mc.mu <- mean(D.mc)        
+    if (mc_style == "scatter") {
         g.mc <- moran_plot(df$Delta, w, xlab = bquote(hat(Delta)))
         } else {    
-            D.mc <- apply(delta.mat, 1, mc, w = w)
-            D.mc.mu <- mean(D.mc)    
             g.mc <- ggplot() +
                 geom_histogram(aes(D.mc),
                                binwidth = binwidth(as.numeric(D.mc)),

README.Rmd

@@ -19,29 +19,38 @@ knitr::opts_chunk$set(
 
 <img src="man/figures/logo.png" align="right" width="160" />
 
-# geostan 
+## geostan: Bayesian spatial analysis
 
 The **geostan** R package supports a complete spatial analysis
 workflow with hierarchical Bayesian models (HBMs) for areal
-data and a variety of functions for visualizing spatial data and model results.
+data, including a variety of functions for visualizing spatial data and model results.
+
+The package is designed primarily to support public health research with spatial data;
+see the [**surveil**](https://connordonegan.github.io/surveil) R package for time series analysis of public health surveillance data.
+
+**geostan** is an interface to [**Stan**](https://mc-stan.org), a state-of-the-art platform for Bayesian inference. 
 
 ### Disease mapping and spatial regression
 
-Model small-area incidence rates with mortality or disease data recorded across areal units like counties or census tracts.
+Model small-area incidence rates with mortality or disease data recorded across areal units like states, counties, or census tracts.
 
 ### Observational uncertainty 
 
-Incorporate information on data reliability into any **geostan** model. Built specifically for American Community Survey (ACS) data.
+Incorporate information on data reliability, such as standard errors of American Community Survey estimates, into any **geostan** model.
+
+### Censored observations
+
+Vital statistics and disease surveillance systems like CDC Wonder censor case counts that fall below a threshold number; **geostan** can provide probability distributions for disease or mortality risk across all areas, censored or not.
 
 ### Spatial analysis tools
 
-Tools for visualizing and measuring spatial autocorrelation and map patterns, for exploratory analysis and model diagnostics.
+Tools for visualizing and measuring spatial autocorrelation and map patterns, for exploratory analysis and model diagnostics. Visual diagnostics also support the evaluation of survey data quality and observational error models.
 
 ### The RStan ecosystem
 
 Compatible with a suite of high-quality R packages for Bayesian inference and model evaluation.
 
-### Custom Stan models
+### Custom spatial models
 
 Tools for building custom spatial models in [Stan](https://mc-stan.org/).
 
@@ -54,26 +63,3 @@ if (!require(drat)) install.packages("drat")
 drat::addRepo("connordonegan")
 install.packages("geostan")
 ```
-
-## Resources
-
-For demonstration analyses and some additional discussion, see the vignettes on the package [website](https://connordonegan.github.io/geostan/) and the package help pages (e.g., run `?stan_car` in R).
-
-## Citation
-
- * Donegan, Connor (2021). geostan: Bayesian Spatial Analysis. R package Version 0.1.1 https://connordonegan.github.io/geostan/
-
- * Donegan, Connor, Yongwan Chun, and Daniel A. Griffith. Modeling community health with areal data: Bayesian inference with survey standard errors and spatial structure. International Journal of Environmental Research and Public Health 18.13 (2021): 6856. DOI: 10.3390/ijerph18136856
-
-
-**geostan** is an interface to **Stan**:
-
- * Carpenter B., Gelman A., Hoffman M. D., Lee D., Goodrich B., Betancourt M., Brubaker M., Guo J., Li P., and Riddell A. (2017). Stan: A probabilistic programming language. Journal of Statistical Software. 76(1). DOI: 10.18637/jss.v076.i01
-
-## Credit
-
-The **geostan** package was built with the help of **rstantools**:
-
- * Gabry, Jonah, Ben Goodrich, and Martin Lysy (2021). rstantools: Tools for Developing R Packages Interfacing with 'Stan'. R package version 2.1.1 https://mc-stan.org/rstantools/index.html
-
-