Handle missing values

Usage

ts_replace_missing(ts, value, indexes = NULL)

ts_interpolate_missing(ts, max_missing = 1, indexes = NULL)

ts_fill_missing_dates(ts, start, end, value = NA)

Arguments

ts: a corehydro_ts.
value: the replacement (or fill) value. ts_fill_missing_dates() defaults to NA, inserting the absent ordinates as missing rather than as zeros.
indexes: optional 1-based ordinate positions to restrict the operation to.
max_missing: the longest run of missing values to interpolate across.
start, end: the date range to fill over.

Value

a corehydro_ts.

Details

ts_replace_missing() sets every missing value to value. ts_interpolate_missing() fills a run of missing values by linear interpolation between its neighbours, but only when the run is no longer than max_missing; a run reaching the END of the series is EXTRAPOLATED from the two preceding ordinates instead. Both interpolate in date space, so an irregular spacing is honoured. ts_fill_missing_dates() is the other half of the problem: it inserts the ordinates a regular series is missing entirely, over the requested date range.

Examples

ts <- time_series(as.Date("2000-01-01") + 0:4, c(3, NA, 4, 1, 5))
ts_interpolate_missing(ts, max_missing = 1)
#> <corehydro_ts> 5 ordinates, interval "one_day"
#>   2000-01-01 to 2000-01-05
#>   2000-01-01  3
#>   2000-01-02  3.5
#>   2000-01-03  4
#>   2000-01-04  1
#>   2000-01-05  5