tvc: Extract temporally-varying coefficients from a fitted model

In tulpa: Templated Unified Library for Posterior Approximation in Bayesian Hierarchical Models

Extract temporally-varying coefficients from a fitted model

Description

Extract posterior distributions of temporally-varying coefficients (TVCs) from a fitted tulpa model with TVC specification.

Usage

tvc(object, terms = NULL, summary = FALSE, probs = c(0.025, 0.5, 0.975), ...)

## S3 method for class 'tulpa_fit'
tvc(object, terms = NULL, summary = FALSE, probs = c(0.025, 0.5, 0.975), ...)

## S3 method for class 'tulpa_tvc_posterior'
print(x, ...)

## S3 method for class 'tulpa_tvc_summary'
print(x, ...)

## S3 method for class 'tulpa_tvc_posterior'
summary(object, probs = c(0.025, 0.5, 0.975), ...)

Arguments

object	A tulpa_fit object fitted with tvc argument
terms	Which TVC terms to extract. If NULL (default), extracts all.
summary	Logical; if TRUE, return summary statistics instead of full posterior draws.
probs	Quantiles to compute if summary = TRUE.
...	Ignored
x	A tulpa_tvc_posterior or tulpa_tvc_summary object.

Value

A tulpa_tvc_posterior object containing:

• draws: Array of posterior draws (draws x times x terms)

• time_levels: Time point labels

• term_names: Names of TVC terms

summary() on a tulpa_tvc_posterior returns a tulpa_tvc_summary data frame with one row per time point and term, holding the posterior mean, SD and requested quantiles of each temporally varying coefficient.

See Also

temporal_tvc(), plot.tulpa_tvc_posterior()

Examples

set.seed(160)
n_t <- 10L; reps <- 5L
walk <- cumsum(rnorm(n_t, 0, 0.35)); walk <- walk - mean(walk)
year <- rep(seq_len(n_t), each = reps)
df <- data.frame(year = year, x = rnorm(length(year)))
df$count <- rpois(nrow(df), exp(0.3 + (0.5 + walk[year]) * df$x))

# The slope on `x` walks in time; TVC is exact-mode only.
fit <- tulpa(
  count ~ x,
  data = df,
  family = "poisson",
  temporal = temporal_tvc("year", terms = ~ x - 1, structure = "rw1"),
  mode = "exact",
  control = list(n_iter = 200L, n_warmup = 100L, seed = 1L)
)

tvc_post <- tvc(fit)
summary(tvc_post)
plot(tvc_post, "x")

tulpa documentation built on Oct. 19, 2026, 5:09 p.m.