Skip to contents

Transition intensities from an msmbayes model, presented as a tidy data frame

Usage

qdf(draws, new_data = NULL, keep_covid = FALSE)

Arguments

draws

Object returned by msmbayes.

new_data

Data frame with covariate values to predict for

keep_covid

(logical) Keep the integer column covid identifying unique covariate combinations.

Value

A data frame with one row per from-state / to-state / covariate value.

Column posterior is in the rvar format of the posterior package, representing a sample from a posterior distribution. Use the summary function on the data frame to produce summary statistics such as the posterior median or mean (see summary.msmbayes).

See also

qmatrix returns the same information in matrix format

Examples

qdf(infsim_model)
#> # A tibble: 2 × 4
#>    from    to    posterior  mode
#>   <int> <int>   <rvar[1d]> <dbl>
#> 1     1     2  0.74 ± 0.34 0.666
#> 2     2     1  4.26 ± 1.98 3.86 
summary(qdf(infsim_model))
#>   from to      mode      mean    median        sd       mad        q5      q95
#> 1    1  2 0.6661191 0.7396011 0.6623418 0.3435138 0.2831664 0.3238362 1.407096
#> 2    2  1 3.8606438 4.2635807 3.8764643 1.9757256 1.6720749 1.8415761 7.972412
#>        rhat ess_bulk ess_tail
#> 1 0.9998919 3891.378 4018.977
#> 2 0.9999342 3893.001 3960.449
summary(qdf(infsim_model), median, ~quantile(.x, c(0.025, 0.975)))
#>   from to      mode    median      2.5%    97.5%
#> 1    1  2 0.6661191 0.6623418 0.2775348 1.618714
#> 2    2  1 3.8606438 3.8764643 1.6053720 9.078582

qdf(infsim_modelc,
    new_data = data.frame(sex=c("female","male")))
#> # A tibble: 4 × 5
#>    from    to    posterior sex     mode
#>   <int> <int>   <rvar[1d]> <chr>  <dbl>
#> 1     1     2  0.69 ± 0.34 female 0.623
#> 2     1     2  0.78 ± 0.37 male   0.704
#> 3     2     1  4.20 ± 1.89 female 3.84 
#> 4     2     1  4.20 ± 1.89 male   3.84