The $summary() method runs
summarise_draws() from the posterior
package and returns the output. For MCMC, only post-warmup draws are
included in the summary.
There is also a $print() method that prints the same summary stats but
removes the extra formatting used for printing tibbles and returns the
fitted model object itself. The $print() method may also be faster than
$summary() because it is designed to only compute the summary statistics
for the variables that will actually fit in the printed output whereas
$summary() will compute them for all of the specified variables in order
to be able to return them to the user. See Examples.
summary(variables = NULL, ...)(character vector) The variables to include.
Optional arguments to pass to posterior::summarise_draws().
The $summary() method returns the tibble data frame created by
posterior::summarise_draws().
The $print() method returns the fitted model object itself (invisibly),
which is the standard behavior for print methods in R.
# \dontrun{
fit <- cmdstanr_example("logistic")
fit$summary()
#> # A tibble: 105 × 10
#> variable mean median sd mad q5 q95 rhat ess_bulk
#> <chr> <num> <num> <num> <num> <num> <num> <num> <num>
#> 1 lp__ -66.0 -65.7 1.45 1.26 -68.9 -64.3 1.00 1950.
#> 2 alpha 0.380 0.376 0.218 0.214 0.0316 0.747 1.00 4585.
#> 3 beta[1] -0.669 -0.662 0.254 0.252 -1.09 -0.248 1.00 4669.
#> 4 beta[2] -0.277 -0.273 0.231 0.232 -0.665 0.0927 1.00 4400.
#> 5 beta[3] 0.676 0.677 0.268 0.267 0.242 1.12 1.00 4179.
#> 6 log_lik[1] -0.515 -0.508 0.0986 0.0967 -0.686 -0.365 1.00 4439.
#> 7 log_lik[2] -0.405 -0.387 0.146 0.140 -0.676 -0.197 1.00 4528.
#> 8 log_lik[3] -0.501 -0.464 0.221 0.207 -0.917 -0.210 1.00 4266.
#> 9 log_lik[4] -0.450 -0.429 0.155 0.148 -0.729 -0.229 1.00 4175.
#> 10 log_lik[5] -1.18 -1.16 0.281 0.278 -1.67 -0.750 1.00 4417.
#> # ℹ 95 more rows
#> # ℹ 1 more variable: ess_tail <num>
fit$print()
#> variable mean median sd mad q5 q95 rhat ess_bulk ess_tail
#> lp__ -66.00 -65.68 1.45 1.26 -68.86 -64.27 1.00 1950 2912
#> alpha 0.38 0.38 0.22 0.21 0.03 0.75 1.00 4584 2789
#> beta[1] -0.67 -0.66 0.25 0.25 -1.09 -0.25 1.00 4669 3265
#> beta[2] -0.28 -0.27 0.23 0.23 -0.67 0.09 1.00 4399 2567
#> beta[3] 0.68 0.68 0.27 0.27 0.24 1.12 1.00 4179 3160
#> log_lik[1] -0.52 -0.51 0.10 0.10 -0.69 -0.37 1.00 4439 2680
#> log_lik[2] -0.40 -0.39 0.15 0.14 -0.68 -0.20 1.00 4528 3426
#> log_lik[3] -0.50 -0.46 0.22 0.21 -0.92 -0.21 1.00 4265 2821
#> log_lik[4] -0.45 -0.43 0.15 0.15 -0.73 -0.23 1.00 4175 2970
#> log_lik[5] -1.18 -1.16 0.28 0.28 -1.67 -0.75 1.00 4417 3002
#>
#> # showing 10 of 105 rows (change via 'max_rows' argument or 'cmdstanr_max_rows' option)
fit$print(max_rows = 2) # same as print(fit, max_rows = 2)
#> variable mean median sd mad q5 q95 rhat ess_bulk ess_tail
#> lp__ -66.00 -65.68 1.45 1.26 -68.86 -64.27 1.00 1950 2912
#> alpha 0.38 0.38 0.22 0.21 0.03 0.75 1.00 4584 2789
#>
#> # showing 2 of 105 rows (change via 'max_rows' argument or 'cmdstanr_max_rows' option)
# include only certain variables
fit$summary("beta")
#> # A tibble: 3 × 10
#> variable mean median sd mad q5 q95 rhat ess_bulk ess_tail
#> <chr> <num> <num> <num> <num> <num> <num> <num> <num> <num>
#> 1 beta[1] -0.669 -0.662 0.254 0.252 -1.09 -0.248 1.00 4669. 3265.
#> 2 beta[2] -0.277 -0.273 0.231 0.232 -0.665 0.0927 1.00 4400. 2568.
#> 3 beta[3] 0.676 0.677 0.268 0.267 0.242 1.12 1.00 4179. 3161.
fit$print(c("alpha", "beta[2]"))
#> variable mean median sd mad q5 q95 rhat ess_bulk ess_tail
#> alpha 0.38 0.38 0.22 0.21 0.03 0.75 1.00 4584 2789
#> beta[2] -0.28 -0.27 0.23 0.23 -0.67 0.09 1.00 4399 2567
# include all variables but only certain summaries
fit$summary(NULL, c("mean", "sd"))
#> # A tibble: 105 × 3
#> variable mean sd
#> <chr> <num> <num>
#> 1 lp__ -66.0 1.45
#> 2 alpha 0.380 0.218
#> 3 beta[1] -0.669 0.254
#> 4 beta[2] -0.277 0.231
#> 5 beta[3] 0.676 0.268
#> 6 log_lik[1] -0.515 0.0986
#> 7 log_lik[2] -0.405 0.146
#> 8 log_lik[3] -0.501 0.221
#> 9 log_lik[4] -0.450 0.155
#> 10 log_lik[5] -1.18 0.281
#> # ℹ 95 more rows
# can use functions created from formulas
# for example, calculate Pr(beta > 0)
fit$summary("beta", prob_gt_0 = ~ mean(. > 0))
#> # A tibble: 3 × 2
#> variable prob_gt_0
#> <chr> <num>
#> 1 beta[1] 0.0035
#> 2 beta[2] 0.112
#> 3 beta[3] 0.995
# can combine user-specified functions with
# the default summary functions
fit$summary(variables = c("alpha", "beta"),
posterior::default_summary_measures()[1:4],
quantiles = ~ quantile2(., probs = c(0.025, 0.975)),
posterior::default_convergence_measures()
)
#> # A tibble: 4 × 10
#> variable mean median sd mad q2.5 q97.5 rhat ess_bulk ess_tail
#> <chr> <num> <num> <num> <num> <num> <num> <num> <num> <num>
#> 1 alpha 0.380 0.376 0.218 0.214 -0.0392 0.823 1.00 4585. 2790.
#> 2 beta[1] -0.669 -0.662 0.254 0.252 -1.18 -0.180 1.00 4669. 3265.
#> 3 beta[2] -0.277 -0.273 0.231 0.232 -0.736 0.165 1.00 4400. 2568.
#> 4 beta[3] 0.676 0.677 0.268 0.267 0.157 1.20 1.00 4179. 3161.
# the functions need to calculate the appropriate
# value for a matrix input
fit$summary(variables = "alpha", dim)
#> # A tibble: 1 × 3
#> variable dim.1 dim.2
#> <chr> <num> <num>
#> 1 alpha 1000 4
# the usual [stats::var()] is therefore not directly suitable as it
# will produce a covariance matrix unless the data is converted to a vector
fit$print(c("alpha", "beta"), var2 = ~var(as.vector(.x)))
#> variable var2
#> alpha 0.05
#> beta[1] 0.06
#> beta[2] 0.05
#> beta[3] 0.07
# }