replace_null_RUV: Fill missing *R*, *U*, and *V* matrices

In MatthewHeun/IEATools: Tools for Working with International Energy Agency Data

Fill missing R, U, and V matrices

Description

In some cases (e.g., bunkers where Last.stage is "final"), R, U, U_feed, U_EIOU, or V matrices can be missing, because imports which appear in the V matrix (or R matrix) are consumed in final demand (Y) matrix, without any intermediate processing. When a data frame is pivoted wider by matrices, the R, U_feed, and U_EIOU columns will contain NULL entries. This function fills those NULL entries with reasonable defaults.

Usage

replace_null_RUV(
  .sutmats = NULL,
  R = IEATools::psut_cols$R,
  U_feed = IEATools::psut_cols$U_feed,
  U_eiou = IEATools::psut_cols$U_eiou,
  U = IEATools::psut_cols$U,
  r_eiou = IEATools::psut_cols$r_eiou,
  Y = IEATools::psut_cols$Y,
  V = IEATools::psut_cols$V,
  resources = IEATools::tpes_flows$resources,
  .R_temp_name = ".R_temp",
  .U_temp_name = ".U_temp",
  .U_feed_temp_name = ".U_feed_temp",
  .U_eiou_temp_name = ".U_EIOU_temp",
  .r_eiou_temp_name = ".r_EIOU_temp",
  .V_temp_name = ".V_temp",
  R_name = IEATools::psut_cols$R,
  U_name = IEATools::psut_cols$U,
  U_feed_name = IEATools::psut_cols$U_feed,
  U_eiou_name = IEATools::psut_cols$U_eiou,
  r_eiou_name = IEATools::psut_cols$r_eiou,
  V_name = IEATools::psut_cols$V
)

Arguments

.sutmats	A data frame of metadata columns and matrix name columns
R, U_feed, U_eiou, U, r_eiou, Y, V	See IEATools::psutcols. Default values are names for variables incoming with .sutmats. Can be overridden with actual matrices.
resources	See IEATools::tpes_flows. The name of the only row of the output 0 R matrix.
.R_temp_name, .U_temp_name, .U_feed_temp_name, .U_eiou_temp_name, .r_eiou_temp_name, .V_temp_name	Names of temporary variables unused internally to the function.
R_name, U_name, U_feed_name, U_eiou_name, r_eiou_name, V_name	See IEATools::psutcols. The final names for matrices in the output.

Details

Reasonable defaults arise from the following thought processes. If all energy is supplied by imports (in the V matrix), there are no resources. Thus, we can replace the missing R matrix with a 0 matrix with a generic "Natural resources" row and the same products as the rows of the Y matrix.

Similarly, missing values for U, U_feed, U_EIOU, or r_EIOU can be replaced by a 0 matrix with row and column names same as a transposed V matrix when it exists. If neither U nor V exist, the R matrix can supply row and column names.

Value

A version of .sutmats with R, U, U_feed, U_EIOU, or V filled with 0 matrices if they were missing.

Examples

# Set up a PSUT data frame with NULL for
# R, U_feed, and U_EIOU in 1971 for GHA.
psut <- load_tidy_iea_df() %>% 
  specify_all() %>% 
  prep_psut() %>% 
  tidyr::pivot_longer(cols = c("R", "U_EIOU", "U_feed", "U", "r_EIOU", "V", "Y", "S_units"),
                      names_to = "matnames", values_to = "matvals") %>% 
  dplyr::filter(!(Country == "GHA" & Year == 1971 & matnames == "R")) %>% 
  dplyr::filter(!(Country == "GHA" & Year == 1971 & matnames == "U_feed")) %>% 
  dplyr::filter(!(Country == "GHA" & Year == 1971 & matnames == "U_EIOU")) %>% 
  tidyr::pivot_wider(names_from = "matnames", values_from = "matvals")
# Replace the `NULL` matrices in the first row.
res <- psut %>% 
  replace_null_RUV()
res$R[[1]]
res$U_feed[[1]]
res$U_EIOU[[1]]

MatthewHeun/IEATools documentation built on Feb. 6, 2024, 3:29 p.m.