Compare Influence Matrices Across Blocks

computes posterior probability of difference between blocks

Usage

compare_blocks(fit, blocks = NULL, parameter = "A", threshold = 0.1)

Arguments

fit

piecewise dbn fit object

blocks

pair of block indices to compare (default: adjacent blocks)

parameter

which parameter to compare: "A" or "B"

threshold

minimum difference to be considered meaningful

Value

list with comparison results

Examples

# \donttest{
sim <- simulate_piecewise_dbn(n = 10, time = 40, blocks = c(15, 30, 40))
fit <- dbn(sim$Y, model = "piecewise", blocks = 3,
           nscan = 200, burn = 100, verbose = FALSE)
compare_blocks(fit)
#> 
#> ── Block Comparison Results (A) 
#> ℹ block_1 vs block_2: ||ΔA|| = 1.895 [1.615, 2.192]
#> →   P(||ΔA|| > 0.1) = 1
#> ℹ block_2 vs block_3: ||ΔA|| = 1.91 [1.628, 2.259]
#> →   P(||ΔA|| > 0.1) = 1
# }