modified best_model function previously using dplyr::filter_() now deprecated to stay on CRAN. Switch to using a dplyr::filter() statement.

Author: lionelvoirol

NAMESPACE

@@ -118,7 +118,7 @@ export(update_obj)
 export(value)
 importFrom(Rcpp,evalCpp)
 importFrom(broom,glance)
-importFrom(dplyr,filter_)
+importFrom(dplyr,filter)
 importFrom(dplyr,group_by)
 importFrom(dplyr,mutate)
 importFrom(grDevices,col2rgb)

R/model_selection.R

@@ -183,23 +183,37 @@ select_ma = function(xt, q.min = 0L, q.max = 3L,
 #'                 q.min = 1L, q.max = 3L)
 #' best_model(x, ic = "hq")
 #' 
-#' @importFrom dplyr filter_
+#' @importFrom dplyr filter
+#' @importFrom magrittr %>%
 best_model = function(x, ic = "aic"){
+
+
+  
+  
 
   criterion = switch(tolower(ic),
                      "aic" = "AIC",
                      "bic" = "BIC",
                      "hq" = "HQ",
                      stop("`criterion` not supported!"))
+
+  o <- x %>%
+    dplyr::filter(ic == criterion, minval)
 
-  crt = paste0("ic == '", criterion,"' & (minval == TRUE)")
+  o$models[[1]]$call$order <- c(o$p[[1]], o$d[[1]], o$q[[1]])
 
-  x %>%
-    filter_(crt) -> o
+  o$models[[1]]
 
-  o$models[[1]]$call$order = eval(parse(text=paste0("c(",o$p,",",o$d,",",o$q,")")))
 
-  o$models[[1]]
+  # remove this due to filter_ is deprecated and filter_ is not working with the new version of dplyr
+  # crt = paste0("ic == '", criterion,"' & (minval == TRUE)")
+  # 
+  # x %>%
+  #   filter_(crt) -> o
+  # 
+  # o$models[[1]]$call$order = eval(parse(text=paste0("c(",o$p,",",o$d,",",o$q,")")))
+  # 
+  # o$models[[1]]
 }