Format Gridded Data

Change the format of a gridded dataset from wide to long (or vice versa) and remove any unmatched area/zones from grid. This is a necessary step for including gridded variables in the conditional logit (logit_c()) model.

Usage

format_grid(
  grid,
  dat,
  project,
  dat.key,
  area.dat,
  area.grid = NULL,
  id.cols,
  from.format = "wide",
  to.format = "wide",
  val.name = NULL,
  save = FALSE
)

Arguments

grid: Gridded dataset to format.
dat: Primary data containing information on hauls or trips. Table in the FishSET database contains the string 'MainDataTable'.
project: Name of project.
dat.key: String, name of column(s) in MainDataTable to join by. The number of columns must match id.cols.
area.dat: String, the name of the area or zone column in dat.
area.grid: String, the name of the area or zone column in dat if from.format = "long". Ignored if from.format = "wide".
id.cols: String, the names of columns from grid that are neither area (area.grid) or value (val.name) columns, for example date or period column(s).
from.format: The original format of grid. Options include "long" or "wide". Use "long" if a single area column exists in grid. Use "wide" if grid contains a column for each area.
to.format: The desired format of grid. Options include "long" or "wide". Use "long" if you want a single area column with a corresponding value column. Use "wide" if you would like each area to have its own column.
val.name: Required if converting from wide to long or long to wide format. When from.format = "wide" and to.format = "long", val.name will be the name of the new value variable associated with the area column. When from.format = "long" and to.format = "wide", val.name will be the name of the existing value variable associated with the area column.
save: Logical, whether to save formatted grid. When TRUE, the table will be saved with the string "Wide" or "Long" appended depending on the value of to.format.

Usage

Arguments

See also