Summarises historical-decomposition contributions over selected periods

Transforms posterior draws of historical decompositions into cumulative structural-shock contributions over a selected historical period and reports their posterior summaries.

Usage

tidy_hd_event(object, ...)

# Default S3 method
tidy_hd_event(object, ...)

# S3 method for class 'bsvar_post_tbl'
tidy_hd_event(
  object,
  start,
  end = start,
  probability = 0.9,
  draws = FALSE,
  ...
)

# S3 method for class 'PosteriorHD'
tidy_hd_event(
  object,
  start,
  end = start,
  probability = 0.9,
  draws = FALSE,
  model = "model1",
  ...
)

# S3 method for class 'PosteriorBSVAR'
tidy_hd_event(
  object,
  start,
  end = start,
  probability = 0.9,
  draws = FALSE,
  model = "model1",
  ...
)

# S3 method for class 'PosteriorBSVARMIX'
tidy_hd_event(
  object,
  start,
  end = start,
  probability = 0.9,
  draws = FALSE,
  model = "model1",
  ...
)

# S3 method for class 'PosteriorBSVARMSH'
tidy_hd_event(
  object,
  start,
  end = start,
  probability = 0.9,
  draws = FALSE,
  model = "model1",
  ...
)

# S3 method for class 'PosteriorBSVARSV'
tidy_hd_event(
  object,
  start,
  end = start,
  probability = 0.9,
  draws = FALSE,
  model = "model1",
  ...
)

# S3 method for class 'PosteriorBSVART'
tidy_hd_event(
  object,
  start,
  end = start,
  probability = 0.9,
  draws = FALSE,
  model = "model1",
  ...
)

# S3 method for class 'PosteriorBSVARSIGN'
tidy_hd_event(
  object,
  start,
  end = start,
  probability = 0.9,
  draws = FALSE,
  model = "model1",
  ...
)

Arguments

object

A posterior model object, PosteriorHD, or a posterior-summary table of historical decompositions.

...

Additional arguments passed to computation methods.

start

First time index to include.

end

Last time index to include. Defaults to start.

probability

Probability mass of the equal-tailed credible interval.

draws

If TRUE, return draw-level cumulative contributions.

model

Label identifying the model specification.

Value

A bsvar_post_tbl with columns model, object_type, variable, shock, event_start, event_end, median, mean, sd, lower, and upper. When draws = TRUE, columns draw and value replace the summary statistics.

See also

posterior_provenance

Examples

data(us_fiscal_lsuw, package = "bsvars")
spec <- bsvars::specify_bsvar$new(us_fiscal_lsuw, p = 1)
#> The identification is set to the default option of lower-triangular structural matrix.
post <- bsvars::estimate(spec, S = 5, show_progress = FALSE)

hd_event <- tidy_hd_event(post, start = "1948.25", end = "1948.5")
head(hd_event)
#> # A tibble: 6 × 21
#>   model  object_type variable shock    mean   median    sd  lower    upper
#>   <chr>  <chr>       <chr>    <chr>   <dbl>    <dbl> <dbl>  <dbl>    <dbl>
#> 1 model1 hd_event    ttr      ttr   -4.24   -5.01    2.21  -5.68  -1.27   
#> 2 model1 hd_event    ttr      gs    -0.0442  0.00234 0.136 -0.228  0.0400 
#> 3 model1 hd_event    ttr      gdp    1.15    1.37    1.15  -0.336  2.25   
#> 4 model1 hd_event    gs       ttr    0.300   0.335   0.450 -0.226  0.805  
#> 5 model1 hd_event    gs       gs     0.0151 -1.33    3.36  -1.97   4.59   
#> 6 model1 hd_event    gs       gdp   -0.104  -0.0466  0.127 -0.247  0.00602
#> # ℹ 12 more variables: analysis_type <chr>, interval_probability <dbl>,
#> #   interval_method <chr>, scale_by <chr>, scale_var <chr>, scale_factor <dbl>,
#> #   units <chr>, source_package <chr>, source_version <chr>, n_draws <int>,
#> #   event_start <chr>, event_end <chr>