Window-based Transition Network Analysis

Computes networks from one-hot (binary indicator) data using temporal windowing. Supports transition (directed), co-occurrence (undirected), or both network types.

Usage

wtna(
  data,
  method = c("transition", "cooccurrence", "both"),
  type = c("frequency", "relative"),
  codes = NULL,
  window_size = 3L,
  mode = c("non-overlapping", "overlapping"),
  actor = NULL
)

Arguments

data: Data frame with one-hot encoded columns (0/1 binary).
method: Character. Network type: "transition" (directed), "cooccurrence" (undirected), or "both" (returns list of two networks). Default: "transition".
type: Character. Output type: "frequency" (raw counts) or "relative" (row-normalized probabilities). Default: "frequency". Note that type = "relative" applied to method = "cooccurrence" produces an asymmetric matrix (conditional co-occurrence given row state), not a symmetric undirected weight matrix - use type = "frequency" if symmetric co-occurrence counts are required.
codes: Character vector or NULL. Names of the one-hot columns to use. If NULL, auto-detects binary columns. Default: NULL.
window_size: Integer (>= 1). Number of consecutive rows to aggregate per window. Default: 3 (windowed pairwise between-window counting). Set window_size = 1 for ordinary consecutive (t -> t+1) transitions with no windowing.
mode: Character. Window mode: "non-overlapping" (fixed, separate windows) or "overlapping" (rolling, step = 1). Default: "non-overlapping".
actor: Character or NULL. Name of the actor/ID column for per-group computation. If NULL, treats all rows as one group. Default: NULL.

Value

For method = "transition" or "cooccurrence": a netobject (see build_network).

For method = "both": a wtna_mixed object with elements $transition and $cooccurrence, each a netobject.

Details

Transitions: Uses crossprod(X[-n,], X[-1,]) to count how often state i is active at time t AND state j at time t+1.

Co-occurrence: Uses crossprod(X) to count states that are simultaneously active in the same row.

Windowing: For window_size > 1, rows are aggregated into windows before computing networks. Non-overlapping windows are fixed, separate blocks; overlapping windows roll forward one row at a time. Within each window, any active indicator (1) in any row makes that state active for the window.

Per-actor: When actor is specified, networks are computed per group and summed.

Examples

oh <- matrix(c(1,0,0, 0,1,0, 0,0,1, 1,0,0), nrow = 4, byrow = TRUE,
             dimnames = list(NULL, c("A","B","C")))
w <- wtna(oh)

# \donttest{
# Simple one-hot data
df <- data.frame(
  A = c(1, 0, 1, 0, 1),
  B = c(0, 1, 0, 1, 0),
  C = c(0, 0, 1, 0, 0)
)

# Transition network
net <- wtna(df)
print(net)
#> Network (method: wtna_transition) [directed]
#>   Weights: [1.000, 2.000]  |  mean: 1.333
#> 
#>   Weight matrix:
#>     A B C
#>   A 2 2 0
#>   B 1 1 0
#>   C 1 1 0 
#> 
#>   Initial probabilities:
#>   A             1.000  ████████████████████████████████████████
#>   B             0.000  
#>   C             0.000  

# Both networks
nets <- wtna(df, method = "both")
print(nets$transition)
#> Network (method: wtna_transition) [directed]
#>   Weights: [1.000, 2.000]  |  mean: 1.333
#> 
#>   Weight matrix:
#>     A B C
#>   A 2 2 0
#>   B 1 1 0
#>   C 1 1 0 
#> 
#>   Initial probabilities:
#>   A             1.000  ████████████████████████████████████████
#>   B             0.000  
#>   C             0.000  
print(nets$cooccurrence)
#> Network (method: wtna_cooccurrence) [undirected]
#>   Weights: [1.000, 5.000]  |  mean: 2.333
#> 
#>   Weight matrix:
#>     A B C
#>   A 5 3 2
#>   B 3 2 1
#>   C 2 1 1 

# With windowing
net <- wtna(df, window_size = 2, mode = "non-overlapping")
# }