some cmdcheck updates; better handling of numeric by variables in gp's

R/globals.R

@@ -20,4 +20,5 @@ utils::globalVariables(c("y", "year", "smooth_vals", "smooth_num",
                          "matches", "time.", "file_name", ".data",
                          "horizon", "target", "Series", "evd", "mean_evd",
                          "total_evd", "smooth_label", "by_variable",
-                         "gr", "tot_subgrs", "subgr"))
+                         "gr", "tot_subgrs", "subgr", "lambda",
+                         "level", "sim_hilbert_gp", "trend_model"))

R/gp.R

@@ -84,7 +84,7 @@ make_gp_additions = function(gp_details,
                                    series = newdata$series,
                                    time = newdata$index..time..index)
     for(i in seq_along(terms_needed)){
-      brms_fake_df_new <- cbind(brms_fake_df_new, data[[terms_needed[i]]])
+      brms_fake_df_new <- cbind(brms_fake_df_new, newdata[[terms_needed[i]]])
     }
     colnames(brms_fake_df_new) <- c('.fake_gp_y',
                                     'series',
@@ -348,7 +348,9 @@ eigenfunc_list = function(stan_data,
         full_eigens[(seq_along(sorted_by))[
           sorted_by == level[x]],] <- eigenfuncs[[x]]
       } else {
-        full_eigens <- eigenfuncs[[x]]
+        # Numeric by variables should be multiplied by the
+        # spectral eigenfunctions
+        full_eigens <- eigenfuncs[[x]] * mock_df[[by[x]]]
       }
     } else {
       full_eigens <- eigenfuncs[[x]]

R/mvgam_formulae.R

@@ -14,16 +14,16 @@
 #' \cr
 #' The formulae supplied to \code{\link{mvgam}} are exactly like those supplied to
 #' \code{\link{glm}} except that smooth terms,
-#' \code{\link[mgcv]{s()}},
-#' \code{\link[mgcv]{te()}},
-#' \code{\link[mgcv]{ti()}} and
-#' \code{\link[mgcv]{t2()}},
-#' time-varying effects using \code{\link{dynamic()}},
+#' \code{\link[mgcv]{s}},
+#' \code{\link[mgcv]{te}},
+#' \code{\link[mgcv]{ti}} and
+#' \code{\link[mgcv]{t2}},
+#' time-varying effects using \code{\link{dynamic}},
 #' monotonically increasing (using `s(x, bs = 'moi')`)
 #' or decreasing splines (using `s(x, bs = 'mod')`;
 #' see \code{\link{smooth.construct.moi.smooth.spec}} for
 #' details), as well as
-#' Gaussian Process functions using \code{\link[brms]{gp()}},
+#' Gaussian Process functions using \code{\link[brms]{gp}},
 #' can be added to the right hand side (and \code{.} is not supported in \code{mvgam} formulae).
 #' \cr
 #' \cr

R/residual_cor.R

@@ -10,31 +10,29 @@
 #' If `TRUE`, the median is used instead. Only used if `summary` is `TRUE`
 #' @param ... ignored
 #' @return If `summary = TRUE`, a `list` with the following components:
-#' \itemize{
-#'  \item{cor, cor_lower, cor_upper}{: A set of \eqn{p \times p} correlation matrices,
+#'  \item{cor, cor_lower, cor_upper}{A set of \eqn{p \times p} correlation matrices,
 #'  containing either the posterior median or mean estimate, plus lower and upper limits
 #'  of the corresponding credible intervals supplied to `probs`}
-#'  \item{sig_cor}{: A \eqn{p \times p} correlation matrix containing only those correlations whose credible
+#'  \item{sig_cor}{A \eqn{p \times p} correlation matrix containing only those correlations whose credible
 #'  interval does not contain zero. All other correlations are set to zero}
-#'  \item{prec, prec_lower, prec_upper}{: A set of \eqn{p \times p} precision matrices,
+#'  \item{prec, prec_lower, prec_upper}{A set of \eqn{p \times p} precision matrices,
 #'  containing either the posterior median or mean estimate, plus lower and upper limits
 #'  of the corresponding credible intervals supplied to `probs`}
-#'  \item{sig_prec}{: A \eqn{p \times p} precision matrix containing only those precisions whose credible
+#'  \item{sig_prec}{A \eqn{p \times p} precision matrix containing only those precisions whose credible
 #'  interval does not contain zero. All other precisions are set to zero}
-#'   \item{cov}{: A \eqn{p \times p} posterior median or mean covariance matrix}
-#'   \item{trace}{: The median/mean point estimator of the trace (sum of the diagonal elements)
+#'   \item{cov}{A \eqn{p \times p} posterior median or mean covariance matrix}
+#'   \item{trace}{The median/mean point estimator of the trace (sum of the diagonal elements)
 #'   of the residual covariance matrix `cov`}
-#'   }
+#'
 #'   If `summary = FALSE`, this function returns a `list` containing the following components:
-#' \itemize{
-#'  \item{all_cormat}{: A \eqn{n_{draws} \times p \times p} `array` of posterior
+#'  \item{all_cormat}{A \eqn{n_{draws} \times p \times p} `array` of posterior
 #'  residual correlation matrix draws}
-#'  \item{all_covmat}{: A \eqn{n_{draws} \times p \times p} `array` of posterior
+#'  \item{all_covmat}{A \eqn{n_{draws} \times p \times p} `array` of posterior
 #'  residual covariance matrix draws}
-#'  \item{all_presmat}{: A \eqn{n_{draws} \times p \times p} `array` of posterior
+#'  \item{all_presmat}{A \eqn{n_{draws} \times p \times p} `array` of posterior
 #'  residual precision matrix draws}
-#'  \item{all_trace}{: A \eqn{n_{draws}} `vector` of posterior covariance trace draws}
-#' }
+#'  \item{all_trace}{A \eqn{n_{draws}} `vector` of posterior covariance trace draws}
+#'
 #' @details
 #' Hui (2016) provides an excellent description of the quantities that this function calculates, so this passage
 #' is heavily paraphrased from his associated `boral` package.

tests/testthat/test-gp.R

@@ -92,8 +92,7 @@ test_that("gp_to_s is working properly for multidimensional gps", {
 test_that("unidimensional gp for observation models working properly", {
   gaus_data$data_train$y[is.na(gaus_data$data_train$y)] <- 0
   mod <- mvgam(formula = y ~ s(series, bs = 're') +
-                 gp(time, by = series, k = 10, c = 5/4,
-                    cov = 'exponential') +
+                 gp(time, by = series, k = 10, c = 5/4) +
                  year:season,
                data = gaus_data$data_train,
                family = gaussian(),