Fit models for use in examples
cmdstanr_example(
example = c("logistic", "schools", "schools_ncp"),
method = c("sample", "optimize", "laplace", "variational", "pathfinder", "diagnose"),
...,
quiet = TRUE,
force_recompile = getOption("cmdstanr_force_recompile", default = FALSE)
)
print_example_program(example = c("logistic", "schools", "schools_ncp"))(string) The name of the example. The currently available examples are
"logistic": logistic regression with intercept and 3 predictors.
"schools": the so-called "eight schools" model, a hierarchical
meta-analysis. Fitting this model will result in warnings about
divergences.
"schools_ncp": non-centered parameterization of the "eight schools"
model that fixes the problem with divergences.
To print the Stan code for a given example use
print_example_program(example).
(string) Which fitting method should be used? The default is
the "sample" method (MCMC).
Arguments passed to the chosen method. See the help pages for
the individual methods for details.
(logical) If TRUE (the default) then fitting the model is
wrapped in utils::capture.output().
Passed to the $compile() method.
The fitted model object returned by the selected method.
# \dontrun{
print_example_program("logistic")
#> data {
#> int<lower=0> N;
#> int<lower=0> K;
#> array[N] int<lower=0, upper=1> y;
#> matrix[N, K] X;
#> }
#> parameters {
#> real alpha;
#> vector[K] beta;
#> }
#> model {
#> target += normal_lpdf(alpha | 0, 1);
#> target += normal_lpdf(beta | 0, 1);
#> target += bernoulli_logit_glm_lpmf(y | X, alpha, beta);
#> }
#> generated quantities {
#> vector[N] log_lik;
#> for (n in 1 : N) {
#> log_lik[n] = bernoulli_logit_lpmf(y[n] | alpha + X[n] * beta);
#> }
#> }
fit_logistic_mcmc <- cmdstanr_example("logistic", chains = 2)
fit_logistic_mcmc$summary()
#> # A tibble: 105 × 10
#> variable mean median sd mad q5 q95 rhat ess_bulk
#> <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 lp__ -65.9 -65.6 1.43 1.23 -68.7 -64.3 1.00 890.
#> 2 alpha 0.377 0.375 0.220 0.220 0.0153 0.743 1.00 2007.
#> 3 beta[1] -0.670 -0.663 0.247 0.244 -1.10 -0.273 1.00 1880.
#> 4 beta[2] -0.268 -0.269 0.223 0.218 -0.637 0.106 1.00 1834.
#> 5 beta[3] 0.684 0.681 0.271 0.271 0.243 1.14 1.00 2163.
#> 6 log_lik[1] -0.516 -0.511 0.101 0.0973 -0.694 -0.363 1.00 2030.
#> 7 log_lik[2] -0.398 -0.38 0.143 0.141 -0.661 -0.204 1.00 2280.
#> 8 log_lik[3] -0.493 -0.463 0.216 0.196 -0.883 -0.197 1.00 1914.
#> 9 log_lik[4] -0.450 -0.430 0.153 0.148 -0.730 -0.229 1.00 2150.
#> 10 log_lik[5] -1.19 -1.16 0.288 0.283 -1.70 -0.769 1.00 2194.
#> # ℹ 95 more rows
#> # ℹ 1 more variable: ess_tail <dbl>
fit_logistic_optim <- cmdstanr_example("logistic", method = "optimize")
fit_logistic_optim$summary()
#> # A tibble: 105 × 2
#> variable estimate
#> <chr> <dbl>
#> 1 lp__ -63.9
#> 2 alpha 0.364
#> 3 beta[1] -0.632
#> 4 beta[2] -0.259
#> 5 beta[3] 0.649
#> 6 log_lik[1] -0.515
#> 7 log_lik[2] -0.394
#> 8 log_lik[3] -0.469
#> 9 log_lik[4] -0.442
#> 10 log_lik[5] -1.14
#> # ℹ 95 more rows
fit_logistic_vb <- cmdstanr_example("logistic", method = "variational")
fit_logistic_vb$summary()
#> # A tibble: 106 × 7
#> variable mean median sd mad q5 q95
#> <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 lp__ -66.2 -65.8 1.56 1.32 -69.2 -64.3
#> 2 lp_approx__ -1.94 -1.61 1.34 1.15 -4.56 -0.362
#> 3 alpha 0.448 0.454 0.184 0.186 0.144 0.745
#> 4 beta[1] -0.710 -0.711 0.297 0.296 -1.20 -0.230
#> 5 beta[2] -0.211 -0.214 0.213 0.218 -0.546 0.131
#> 6 beta[3] 0.719 0.717 0.272 0.278 0.285 1.18
#> 7 log_lik[1] -0.478 -0.474 0.0839 0.0865 -0.622 -0.351
#> 8 log_lik[2] -0.391 -0.364 0.154 0.141 -0.698 -0.180
#> 9 log_lik[3] -0.408 -0.382 0.181 0.171 -0.755 -0.162
#> 10 log_lik[4] -0.483 -0.471 0.145 0.142 -0.745 -0.272
#> # ℹ 96 more rows
print_example_program("schools")
#> data {
#> int<lower=1> J;
#> vector<lower=0>[J] sigma;
#> vector[J] y;
#> }
#> parameters {
#> real mu;
#> real<lower=0> tau;
#> vector[J] theta;
#> }
#> model {
#> target += normal_lpdf(tau | 0, 10);
#> target += normal_lpdf(mu | 0, 10);
#> target += normal_lpdf(theta | mu, tau);
#> target += normal_lpdf(y | theta, sigma);
#> }
fit_schools_mcmc <- cmdstanr_example("schools")
#> Warning: 679 of 4000 (17.0%) transitions ended with a divergence.
#> See https://mc-stan.org/misc/warnings for details.
#> Warning: 2 of 4 chains had an E-BFMI less than 0.3.
#> See https://mc-stan.org/misc/warnings for details.
fit_schools_mcmc$summary()
#> # A tibble: 11 × 10
#> variable mean median sd mad q5 q95 rhat ess_bulk ess_tail
#> <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 lp__ -55.2 -56.2 7.20 8.44 -65.7 -44.5 1.35 9.26 502.
#> 2 mu 5.92 4.83 3.74 2.79 0.0739 12.5 1.14 269. 979.
#> 3 tau 4.29 3.34 3.72 3.44 0.650 11.7 1.35 9.24 5.53
#> 4 theta[1] 8.28 6.13 6.22 3.70 0.148 20.3 1.19 199. 1448.
#> 5 theta[2] 6.41 5.49 5.11 3.45 -1.72 15.4 1.10 460. 1260.
#> 6 theta[3] 5.14 4.33 5.96 3.94 -5.01 15.0 1.23 782. 1447.
#> 7 theta[4] 5.87 4.59 5.16 3.49 -2.48 15.2 1.31 730. 948.
#> 8 theta[5] 4.56 4.37 5.04 4.04 -4.72 12.5 1.28 643. 1270.
#> 9 theta[6] 5.08 4.56 5.20 3.35 -3.77 13.6 1.12 659. 1302.
#> 10 theta[7] 8.06 6.46 5.45 3.53 0.928 18.2 1.05 73.8 1255.
#> 11 theta[8] 6.38 5.50 6.15 3.54 -2.77 17.5 1.11 260. 1699.
print_example_program("schools_ncp")
#> data {
#> int<lower=1> J;
#> vector<lower=0>[J] sigma;
#> vector[J] y;
#> }
#> parameters {
#> real mu;
#> real<lower=0> tau;
#> vector[J] theta_raw;
#> }
#> transformed parameters {
#> vector[J] theta = mu + tau * theta_raw;
#> }
#> model {
#> target += normal_lpdf(tau | 0, 10);
#> target += normal_lpdf(mu | 0, 10);
#> target += normal_lpdf(theta_raw | 0, 1);
#> target += normal_lpdf(y | theta, sigma);
#> }
fit_schools_ncp_mcmc <- cmdstanr_example("schools_ncp")
fit_schools_ncp_mcmc$summary()
#> # A tibble: 19 × 10
#> variable mean median sd mad q5 q95 rhat ess_bulk
#> <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 lp__ -46.9 -46.6 2.41 2.29 -51.3 -43.5 1.00 1454.
#> 2 mu 6.51 6.48 4.11 4.18 -0.0368 13.3 1.00 3392.
#> 3 tau 4.77 3.96 3.63 3.48 0.449 11.8 1.00 2260.
#> 4 theta_raw[1] 0.377 0.379 0.975 0.954 -1.26 1.95 1.00 4065.
#> 5 theta_raw[2] 0.0526 0.0555 0.895 0.884 -1.43 1.53 1.00 3436.
#> 6 theta_raw[3] -0.143 -0.156 0.970 0.957 -1.73 1.48 1.00 3686.
#> 7 theta_raw[4] 0.0219 0.0261 0.906 0.930 -1.48 1.50 1.00 3671.
#> 8 theta_raw[5] -0.263 -0.285 0.920 0.904 -1.76 1.29 1.00 4086.
#> 9 theta_raw[6] -0.130 -0.144 0.930 0.917 -1.64 1.42 1.00 4353.
#> 10 theta_raw[7] 0.369 0.376 0.918 0.911 -1.15 1.87 1.00 3860.
#> 11 theta_raw[8] 0.0659 0.0592 0.971 0.988 -1.51 1.69 1.00 3781.
#> 12 theta[1] 9.01 8.30 6.74 5.76 -0.246 21.2 1.00 3647.
#> 13 theta[2] 6.88 6.83 5.35 4.89 -1.89 15.7 1.00 4377.
#> 14 theta[3] 5.45 5.79 6.51 5.55 -5.90 15.5 1.00 3662.
#> 15 theta[4] 6.73 6.60 5.73 5.22 -2.58 16.1 1.00 3999.
#> 16 theta[5] 4.91 5.16 5.52 4.90 -4.78 13.6 1.00 4517.
#> 17 theta[6] 5.64 5.91 6.01 5.33 -4.77 15.0 1.00 4197.
#> 18 theta[7] 8.87 8.36 5.95 5.37 0.312 19.6 1.00 3846.
#> 19 theta[8] 7.06 6.77 6.80 5.70 -3.70 18.3 1.00 3866.
#> # ℹ 1 more variable: ess_tail <dbl>
# optimization fails for hierarchical model
cmdstanr_example("schools", "optimize", quiet = FALSE)
#> Initial log joint probability = -53.966
#> Iter log prob ||dx|| ||grad|| alpha alpha0 # evals Notes
#> 99 123.351 0.058381 2.09576e+09 0.5049 0.5049 181
#> Iter log prob ||dx|| ||grad|| alpha alpha0 # evals Notes
#> 184 248.826 0.4674 3.92933e+16 1e-12 0.001 384 LS failed, Hessian reset
#> Chain 1 Optimization terminated with error:
#> Chain 1 Line search failed to achieve a sufficient decrease, no more progress can be made
#> Warning: Fitting finished unexpectedly! Use the $output() method for more information.
#> Finished in 0.2 seconds.
#> Error: Fitting failed. Unable to print.
# }