Per-context table of a path-dependence analysis

Usage

# S3 method for class 'net_path_dependence'
hg_get(
  x,
  what = "contexts",
  ...,
  min_count = NULL,
  flips = NULL,
  sort_by = "kl",
  top = NULL
)

Arguments

x: A net_path_dependence object from memory().
what: "contexts", the only table.
...: Additional arguments (ignored).
min_count: Integer or NULL. Keep only contexts observed at least this many times.
flips: NULL (default, every context), TRUE for only the contexts whose most probable next state differs from the first-order prediction (flips is TRUE), or FALSE for only those where it agrees.
sort_by: "kl" (default) sorts by the order-k vs order-1 divergence, largest first; NULL keeps the construction order.
top: Integer or NULL. Return only the first top rows, applied after any filter and after sort_by, so sort_by and top compose. Default NULL returns every row.

Value

A data.frame, one row per context: context, count (times observed), entropy_first_order and entropy_order_k (entropy of the next state given the last state and given the whole context), entropy_drop (their difference), kl (Kullback-Leibler divergence of the order-k from the first-order distribution), top_first_order and top_order_k (the most probable next state under each), and flips (TRUE when the two differ).

Examples

# one sequence per row
wide <- data.frame(t1 = c("a", "x", "a", "x"), t2 = c("b", "b", "b", "b"),
                   t3 = c("c", "d", "c", "d"), t4 = c("a", "x", "a", "x"),
                   t5 = c("b", "b", "b", "b"), t6 = c("c", "d", "c", "d"))
pd <- memory(wide, order = 2, min_count = 1)
hg_get(pd)
#>   context count entropy_first_order entropy_order_k entropy_drop kl
#> 1  a -> b     4                   1               0            1  1
#> 2  x -> b     4                   1               0            1  1
#> 3  b -> c     2                   0               0            0  0
#> 4  b -> d     2                   0               0            0  0
#> 5  c -> a     2                   0               0            0  0
#> 6  d -> x     2                   0               0            0  0
#>   top_first_order top_order_k flips
#> 1               c           c FALSE
#> 2               c           d  TRUE
#> 3               a           a FALSE
#> 4               x           x FALSE
#> 5               b           b FALSE
#> 6               b           b FALSE
hg_get(pd, flips = TRUE)
#>   context count entropy_first_order entropy_order_k entropy_drop kl
#> 1  x -> b     4                   1               0            1  1
#>   top_first_order top_order_k flips
#> 1               c           d  TRUE