Fitted values on the training design

fitted.mixgpd_fit() is a thin training-data wrapper around predict.mixgpd_fit for conditional models.

Usage

# S3 method for class 'mixgpd_fit'
fitted(
  object,
  type = c("mean", "median", "quantile"),
  p = 0.5,
  level = 0.95,
  interval = "credible",
  seed = 1,
  ...
)

Arguments

object

A fitted object of class "mixgpd_fit" (must have covariates).

type

Which fitted functional to return:

"mean": posterior predictive mean
"median": posterior predictive median
"quantile": posterior predictive quantile at level p

p

Quantile level used when type = "quantile".

level

Credible level for confidence intervals (default 0.95 for 95 percent credible intervals).

interval

Character or NULL; type of credible interval:

NULL: no interval
"credible" (default): equal-tailed quantile intervals
"hpd": highest posterior density intervals

seed

Random seed used for deterministic fitted values.

...

Unused.

Value

A data frame with columns for fitted values, optional intervals, and residuals computed on the training sample.

Details

The method returns posterior predictive fitted values on the observed design matrix. It is available only when the fitted model stored covariates.

Examples

# \donttest{
# Conditional model (with covariates X)
y <- abs(stats::rnorm(25)) + 0.1
X <- data.frame(x1 = stats::rnorm(25), x2 = stats::runif(25))
bundle <- build_nimble_bundle(y = y, X = X, backend = "sb", kernel = "normal",
                             GPD = TRUE, components = 3,
                             mcmc = list(niter = 100, nburnin = 50, thin = 1, nchains = 1))
fit <- run_mcmc_bundle_manual(bundle)
#> [mixgpd] Validating configuration
#> [mixgpd] Checking build/compile cache
#> [mixgpd] Building model and MCMC configuration
#> [mixgpd] Compiling NIMBLE model
#> [mixgpd] Initializing chains
#> [mixgpd] Running MCMC
#> [mixgpd] Finalizing WAIC and diagnostics
#> [mixgpd] Assembling fit object
fitted(fit)
#> [predict_mixgpd] Validating prediction inputs
#> [predict_mixgpd] Extracting posterior draws
#> [predict_mixgpd] Computing posterior summaries
#> [predict_mixgpd] Assembling prediction output
#>             fit      lower      upper residuals
#> 1  -0.137844708 -0.4015448 0.12543383 0.6670157
#> 2  -0.107184615 -0.3327693 0.08011055 0.4288305
#> 3  -0.001528761 -0.2427112 0.32436075 0.7843265
#> 4  -0.036011726 -0.2195284 0.09354108 0.4781796
#> 5  -0.114233327 -0.3587995 0.13567649 1.2213985
#> 6  -0.072555424 -0.2588915 0.06448799 1.1210418
#> 7  -0.099940952 -0.3180770 0.07542794 0.4245665
#> 8  -0.004311994 -0.2035797 0.17775033 2.2871491
#> 9  -0.146150363 -0.4210200 0.12102086 1.3093515
#> 10 -0.060774856 -0.2973570 0.29927904 0.8438332
#> 11 -0.117446843 -0.3345290 0.11191342 0.8604083
#> 12  0.021507282 -0.2507034 0.46647472 1.9157100
#> 13 -0.086479196 -0.2905966 0.07158354 0.8812904
#> 14 -0.115478340 -0.3326636 0.09821612 1.2918690
#> 15 -0.137440502 -0.3902392 0.11076055 0.6966442
#> 16 -0.063630012 -0.2499352 0.05374555 1.3035068
#> 17 -0.130091725 -0.3670486 0.11064810 0.3687356
#> 18 -0.134228569 -0.3775809 0.11589957 1.4664903
#> 19 -0.024279756 -0.2142548 0.11993065 1.9547894
#> 20 -0.093428157 -0.2898281 0.12885299 0.5118388
#> 21 -0.139103707 -0.4164418 0.10523679 1.0129575
#> 22 -0.125623149 -0.3842722 0.11998722 0.8552477
#> 23 -0.080243723 -0.2707207 0.13855523 0.9483501
#> 24 -0.137968807 -0.4042290 0.12083928 1.0510353
#> 25 -0.033843873 -0.2477778 0.17686425 1.3976087
fitted(fit, level = 0.90)
#> [predict_mixgpd] Validating prediction inputs
#> [predict_mixgpd] Extracting posterior draws
#> [predict_mixgpd] Computing posterior summaries
#> [predict_mixgpd] Assembling prediction output
#>             fit      lower      upper residuals
#> 1  -0.137844708 -0.3471485 0.06383905 0.6670157
#> 2  -0.107184615 -0.2861716 0.06652476 0.4288305
#> 3  -0.001528761 -0.2141012 0.26869621 0.7843265
#> 4  -0.036011726 -0.2076551 0.08498833 0.4781796
#> 5  -0.114233327 -0.3113510 0.05231331 1.2213985
#> 6  -0.072555424 -0.2437264 0.04989450 1.1210418
#> 7  -0.099940952 -0.2699493 0.05255102 0.4245665
#> 8  -0.004311994 -0.1794000 0.15801906 2.2871491
#> 9  -0.146150363 -0.3627920 0.06884455 1.3093515
#> 10 -0.060774856 -0.2932251 0.19378313 0.8438332
#> 11 -0.117446843 -0.3178055 0.03235647 0.8604083
#> 12  0.021507282 -0.2183117 0.36791304 1.9157100
#> 13 -0.086479196 -0.2464349 0.05961112 0.8812904
#> 14 -0.115478340 -0.3093543 0.01791537 1.2918690
#> 15 -0.137440502 -0.3468084 0.03329134 0.6966442
#> 16 -0.063630012 -0.2285502 0.04998070 1.3035068
#> 17 -0.130091725 -0.3361768 0.02008063 0.3687356
#> 18 -0.134228569 -0.3437525 0.02463583 1.4664903
#> 19 -0.024279756 -0.1944556 0.10550474 1.9547894
#> 20 -0.093428157 -0.2838919 0.08723265 0.5118388
#> 21 -0.139103707 -0.3595531 0.09291596 1.0129575
#> 22 -0.125623149 -0.3332913 0.06855453 0.8552477
#> 23 -0.080243723 -0.2627646 0.08546870 0.9483501
#> 24 -0.137968807 -0.3496137 0.07209921 1.0510353
#> 25 -0.033843873 -0.2319713 0.15794859 1.3976087
fitted(fit, interval = "hpd")  # HPD intervals
#> [predict_mixgpd] Validating prediction inputs
#> [predict_mixgpd] Extracting posterior draws
#> [predict_mixgpd] Computing posterior summaries
#> [predict_mixgpd] Assembling prediction output
#>             fit      lower      upper residuals
#> 1  -0.137844708 -0.4288305 0.13253502 0.6670157
#> 2  -0.107184615 -0.3649412 0.08318867 0.4288305
#> 3  -0.001528761 -0.2675449 0.33569199 0.7843265
#> 4  -0.036011726 -0.2216595 0.09526185 0.4781796
#> 5  -0.114233327 -0.4130114 0.15983090 1.2213985
#> 6  -0.072555424 -0.2853117 0.06640453 1.1210418
#> 7  -0.099940952 -0.3340408 0.07748816 0.4245665
#> 8  -0.004311994 -0.2229704 0.17966263 2.2871491
#> 9  -0.146150363 -0.4374930 0.13077611 1.3093515
#> 10 -0.060774856 -0.3363083 0.32534781 0.8438332
#> 11 -0.117446843 -0.3859923 0.13489128 0.8604083
#> 12  0.021507282 -0.2842261 0.49163802 1.9157100
#> 13 -0.086479196 -0.3146479 0.07372831 0.8812904
#> 14 -0.115478340 -0.3760022 0.11953940 1.2918690
#> 15 -0.137440502 -0.4177385 0.13149828 0.6966442
#> 16 -0.063630012 -0.2649798 0.05422633 1.3035068
#> 17 -0.130091725 -0.4057259 0.13625130 0.3687356
#> 18 -0.134228569 -0.4145869 0.14166964 1.4664903
#> 19 -0.024279756 -0.2175516 0.12358693 1.9547894
#> 20 -0.093428157 -0.3431809 0.13326052 0.5118388
#> 21 -0.139103707 -0.4647031 0.10674876 1.0129575
#> 22 -0.125623149 -0.4346648 0.13405370 0.8552477
#> 23 -0.080243723 -0.2718122 0.17745069 0.9483501
#> 24 -0.137968807 -0.4394089 0.12870719 1.0510353
#> 25 -0.033843873 -0.2520106 0.17902425 1.3976087
fitted(fit, interval = NULL)   # No intervals
#> [predict_mixgpd] Validating prediction inputs
#> [predict_mixgpd] Extracting posterior draws
#> [predict_mixgpd] Computing posterior summaries
#> [predict_mixgpd] Assembling prediction output
#>             fit lower upper residuals
#> 1  -0.137844708    NA    NA 0.6670157
#> 2  -0.107184615    NA    NA 0.4288305
#> 3  -0.001528761    NA    NA 0.7843265
#> 4  -0.036011726    NA    NA 0.4781796
#> 5  -0.114233327    NA    NA 1.2213985
#> 6  -0.072555424    NA    NA 1.1210418
#> 7  -0.099940952    NA    NA 0.4245665
#> 8  -0.004311994    NA    NA 2.2871491
#> 9  -0.146150363    NA    NA 1.3093515
#> 10 -0.060774856    NA    NA 0.8438332
#> 11 -0.117446843    NA    NA 0.8604083
#> 12  0.021507282    NA    NA 1.9157100
#> 13 -0.086479196    NA    NA 0.8812904
#> 14 -0.115478340    NA    NA 1.2918690
#> 15 -0.137440502    NA    NA 0.6966442
#> 16 -0.063630012    NA    NA 1.3035068
#> 17 -0.130091725    NA    NA 0.3687356
#> 18 -0.134228569    NA    NA 1.4664903
#> 19 -0.024279756    NA    NA 1.9547894
#> 20 -0.093428157    NA    NA 0.5118388
#> 21 -0.139103707    NA    NA 1.0129575
#> 22 -0.125623149    NA    NA 0.8552477
#> 23 -0.080243723    NA    NA 0.9483501
#> 24 -0.137968807    NA    NA 1.0510353
#> 25 -0.033843873    NA    NA 1.3976087
# }

Usage

Arguments

Value

Details

See also

Examples