Impute panel data — impute

Given a data frame of panel data, with a time-index column time_col and a unit ID column unit_col, imputes other columns using the entry from the latest available time point.

Usage

impute_panel(
  iData,
  time_col = NULL,
  unit_col = NULL,
  cols = NULL,
  imp_type = NULL,
  max_time = NULL
)

Arguments

iData: A data frame of indicator data, containing a time index column time_col, a unit code column unit_col, and other numerical columns to be imputed.
time_col: The name of a column found in iData to be used as the time index column. Must point to a numeric column.
unit_col: The name of a column found in iData to be used as the unit code/ID column. Must point to a character column.
cols: Optionally, a character vector of names of columns to impute. If NULL (default), all columns apart from time_col and unit_col will be imputed where possible.
imp_type: One of "latest" "constant", "linear" or "linear-constant". In the first case, missing points are imputed with the last non-NA observation for each time series, up to max_time. For "constant" or "linear", missing points are imputed using stats::approx(), passing "constant" or "linear" to the method argument, and points outside of the range of observed values are replaced with the nearest non-NA point. This is equivalent to rule = 2 in stats::approx() for each time series. The difference between "latest" and "constant" is that the latter allows control over the maximum number of time points to impute backwards (using max_time) whereas the former doesn't. Additionally, "constant" will impute outside of the observed range of values at the beginning of the time series, whereas "latest" won't. Finally, the "linear-constant" option will apply linear imputation where possible, but will revert to the "constant" method for any time series with only one observation, which would otherwise throw an error for "linear".
max_time: The maximum number of time points to look backwards to impute from. E.g. if max_time = 1, if an NA is found at time $t$, it will only look for a replacement value at $t-1$ but not in any time points before that. By default, searches all time points available.

Value

A list containing:

.$iData_imp: An iData format data frame with missing data imputed using previous time points (where possible).
.$DataT: A data frame in the same format as iData, where each entry shows which time point each data point came from.

Details

This presumes that there are multiple observations for each unit code, i.e. one per time point. It then searches for any missing values in the target year, and replaces them with the equivalent points from previous time points. It will replace using the most recently available point or using linear interpolation: see imp_type argument.

Examples

# Copy example panel data
iData_p <- ASEM_iData_p

# we introduce two NAs: one for NZ in 2022 in LPI indicator
iData_p$LPI[iData_p$uCode == "NZ" & iData_p$Time == 2022] <-  NA
# one for AT, also in 2022, but for Flights indicator
iData_p$Flights[iData_p$uCode == "AT" & iData_p$Time == 2022] <- NA

# impute: target only the two columns where NAs introduced
l_imp <- impute_panel(iData_p, cols = c("LPI", "Flights"))
# get imputed df
iData_imp <- l_imp$iData_imp

# check the output is what we expect: both NAs introduced should now have 2021 values
iData_imp$LPI[iData_imp$uCode == "NZ" & iData_imp$Time == 2022] ==
  ASEM_iData_p$LPI[ASEM_iData_p$uCode == "NZ" & ASEM_iData_p$Time == 2021]
#> logical(0)

iData_imp$Flights[iData_imp$uCode == "AT" & iData_imp$Time == 2022] ==
  ASEM_iData_p$Flights[ASEM_iData_p$uCode == "AT" & ASEM_iData_p$Time == 2021]
#> logical(0)