Derive a sequence data structure from episode data

seqpe: builds a sequence – a state at every time point – from episode data (a start time, an end time, and origin/destination states for each spell). The result can be read back with a fresh call to tda_seqm or any of the other sequence functions once written out and re-read; this function only performs the conversion, since TDA's output is a plain data file rather than something kept in memory across calls the way tda_seqm manages internally.

Usage

tda_seqpe(
  data,
  id,
  origin,
  destination,
  start,
  end,
  tp,
  missing = NULL,
  missing_start = NULL,
  missing_end = NULL,
  options = list(),
  dir = tempfile("tda"),
  ...
)

Arguments

data

a data frame of episodes, one row per spell.

id

name of the case identifier column in data.

origin, destination

names of the origin- and destination-state columns.

start, end

names of the start- and end-time columns.

tp

definition of the time points the sequence is built over, as a vector or a single TDA range expression such as "0(1)10".

missing

value to use for a gap at the beginning (missing_start), at the end (missing_end), or anywhere else (missing); TDA's default is -1 for all three.

missing_start, missing_end

see missing.

options

a named list of further TDA options, passed through.

dir

working directory.

...

passed to tda_run.

Value

A data frame: the case id, then one column per time point in tp, each holding the state the case was in at that time.

See also

Other sequence analysis: tda_evalfi(), tda_fml(), tda_frml(), tda_seq_info(), tda_seqgc(), tda_seqm(), tda_seqmd(), tda_seqpm()

Examples

# two cases, two spells each
e <- data.frame(id = c(1, 1, 2, 2), org = c(1, 2, 1, 3),
                des = c(2, 3, 3, 1), ts = c(0, 3, 0, 2),
                tf = c(3, 5, 2, 5))
tda_seqpe(e, id = "id", origin = "org", destination = "des",
         start = "ts", end = "tf", tp = "0(1)5")
#>   id t0 t1 t2 t3 t4 t5
#> 1  1  1  1  1  2  2 -1
#> 2  2  1  1  3  3  3 -1