diff --git a/R/forecast.mvgam.R b/R/forecast.mvgam.R index 2c31790e..eec631c3 100644 --- a/R/forecast.mvgam.R +++ b/R/forecast.mvgam.R @@ -34,7 +34,8 @@ forecast <- function(object, ...){ #' trend_model = AR(), #' noncentred = TRUE, #' data = simdat$data_train, -#' chains = 2) +#' chains = 2, +#' silent = 2) #' #' # Hindcasts on response scale #' hc <- hindcast(mod) @@ -864,7 +865,8 @@ forecast_draws = function(object, 'linkfun', 'forecast_trend', 'extract_series_trend_pars', - 'mvgam_predict') + 'mvgam_predict', + 'prep_varma_params') attr(funs_list, 'envir') <- as.environment(asNamespace("mvgam")) attr(funs_list, 'mode') <- 'function' diff --git a/R/lfo_cv.mvgam.R b/R/lfo_cv.mvgam.R index bb3ee157..4a9655ed 100644 --- a/R/lfo_cv.mvgam.R +++ b/R/lfo_cv.mvgam.R @@ -70,7 +70,8 @@ #' newdata = dat$data_test, #' burnin = 300, #' samples = 300, -#' chains = 2) +#' chains = 2, +#' silent = 2) #' #'# Fit a less appropriate model #'mod_rw <- mvgam(y ~ s(season, bs = 'cc', k = 6), @@ -80,7 +81,8 @@ #' newdata = dat$data_test, #' burnin = 300, #' samples = 300, -#' chains = 2) +#' chains = 2, +#' silent = 2) #' #'# Compare Discrete Ranked Probability Scores for the testing period #'fc_ar2 <- forecast(mod_ar2) @@ -96,10 +98,12 @@ #'# for estimating model parameters #'lfo_ar2 <- lfo_cv(mod_ar2, #' min_t = 40, -#' fc_horizon = 3) +#' fc_horizon = 3, +#' silent = 2) #'lfo_rw <- lfo_cv(mod_rw, #' min_t = 40, -#' fc_horizon = 3) +#' fc_horizon = 3, +#' silent = 2) #' #'# Plot Pareto-K values and ELPD estimates #'plot(lfo_ar2) diff --git a/R/loo.mvgam.R b/R/loo.mvgam.R index 9e0a85be..f98cdec2 100644 --- a/R/loo.mvgam.R +++ b/R/loo.mvgam.R @@ -52,7 +52,8 @@ #' data = rbind(simdat$data_train, #' simdat$data_test), #' family = gaussian(), -#' chains = 2) +#' chains = 2, +#' silent = 2) #' #'# Inspect the model and calculate LOO #'conditional_effects(mod1) @@ -65,14 +66,16 @@ #' formula = y ~ s(season, bs = 'cc', k = 6) + #' s(season, series, bs = 'fs', #' xt = list(bs = 'cc'), k = 4), -#' chains = 2) +#' chains = 2, +#' silent = 2) #'conditional_effects(mod2) #'loo(mod2) #' #'# Now add AR1 dynamic errors to mod2 #'mod3 <- update(mod2, #' trend_model = AR(), -#' chains = 2) +#' chains = 2, +#' silent = 2) #'conditional_effects(mod3) #'plot(mod3, type = 'trend') #'loo(mod3) diff --git a/docs/reference/forecast.mvgam-1.png b/docs/reference/forecast.mvgam-1.png index 47e96a3b..2be6e7cb 100644 Binary files a/docs/reference/forecast.mvgam-1.png and b/docs/reference/forecast.mvgam-1.png differ diff --git a/docs/reference/forecast.mvgam-10.png b/docs/reference/forecast.mvgam-10.png index a83dca05..67cc7dde 100644 Binary files a/docs/reference/forecast.mvgam-10.png and b/docs/reference/forecast.mvgam-10.png differ diff --git a/docs/reference/forecast.mvgam-11.png b/docs/reference/forecast.mvgam-11.png index a7829fdb..1dcf44cd 100644 Binary files a/docs/reference/forecast.mvgam-11.png and b/docs/reference/forecast.mvgam-11.png differ diff --git a/docs/reference/forecast.mvgam-12.png b/docs/reference/forecast.mvgam-12.png index 0a8e58a8..19157c19 100644 Binary files a/docs/reference/forecast.mvgam-12.png and b/docs/reference/forecast.mvgam-12.png differ diff --git a/docs/reference/forecast.mvgam-2.png b/docs/reference/forecast.mvgam-2.png index e04aac1b..cffa10db 100644 Binary files a/docs/reference/forecast.mvgam-2.png and b/docs/reference/forecast.mvgam-2.png differ diff --git a/docs/reference/forecast.mvgam-3.png b/docs/reference/forecast.mvgam-3.png index 1343af62..302c192a 100644 Binary files a/docs/reference/forecast.mvgam-3.png and b/docs/reference/forecast.mvgam-3.png differ diff --git a/docs/reference/forecast.mvgam-4.png b/docs/reference/forecast.mvgam-4.png index 3b89a4c1..40780f73 100644 Binary files a/docs/reference/forecast.mvgam-4.png and b/docs/reference/forecast.mvgam-4.png differ diff --git a/docs/reference/forecast.mvgam-5.png b/docs/reference/forecast.mvgam-5.png index b8986778..c0dbf906 100644 Binary files a/docs/reference/forecast.mvgam-5.png and b/docs/reference/forecast.mvgam-5.png differ diff --git a/docs/reference/forecast.mvgam-6.png b/docs/reference/forecast.mvgam-6.png index 9a8f2091..e1d3bb89 100644 Binary files a/docs/reference/forecast.mvgam-6.png and b/docs/reference/forecast.mvgam-6.png differ diff --git a/docs/reference/forecast.mvgam-7.png b/docs/reference/forecast.mvgam-7.png index c7a603a5..32a5e315 100644 Binary files a/docs/reference/forecast.mvgam-7.png and b/docs/reference/forecast.mvgam-7.png differ diff --git a/docs/reference/forecast.mvgam-8.png b/docs/reference/forecast.mvgam-8.png index 22f8e190..90f4d41a 100644 Binary files a/docs/reference/forecast.mvgam-8.png and b/docs/reference/forecast.mvgam-8.png differ diff --git a/docs/reference/forecast.mvgam-9.png b/docs/reference/forecast.mvgam-9.png index a7089b6c..f3b2705d 100644 Binary files a/docs/reference/forecast.mvgam-9.png and b/docs/reference/forecast.mvgam-9.png differ diff --git a/docs/reference/forecast.mvgam.html b/docs/reference/forecast.mvgam.html index 4440451f..ef9f5a10 100644 --- a/docs/reference/forecast.mvgam.html +++ b/docs/reference/forecast.mvgam.html @@ -141,9 +141,8 @@

Examples trend_model = AR(), noncentred = TRUE, data = simdat$data_train, - chains = 2) -#> Compiling Stan program using cmdstanr -#> + chains = 2, + silent = 2) #> In file included from stan/lib/stan_math/stan/math/prim/prob/von_mises_lccdf.hpp:5, #> from stan/lib/stan_math/stan/math/prim/prob/von_mises_ccdf_log.hpp:4, #> from stan/lib/stan_math/stan/math/prim/prob.hpp:359, @@ -151,53 +150,19 @@

Examples#> from stan/lib/stan_math/stan/math/rev.hpp:16, #> from stan/lib/stan_math/stan/math.hpp:19, #> from stan/src/stan/model/model_header.hpp:4, -#> from C:/Users/uqnclar2/AppData/Local/Temp/RtmpARA4r4/model-1333c59e1e97.hpp:2: +#> from C:/Users/uqnclar2/AppData/Local/Temp/Rtmp0U8hqj/model-20b0383b52a8.hpp:2: #> stan/lib/stan_math/stan/math/prim/prob/von_mises_cdf.hpp: In function 'stan::return_type_t<T_x, T_sigma, T_l> stan::math::von_mises_cdf(const T_x&, const T_mu&, const T_k&)': #> stan/lib/stan_math/stan/math/prim/prob/von_mises_cdf.hpp:194: note: '-Wmisleading-indentation' is disabled from this point onwards, since column-tracking was disabled due to the size of the code/headers #> 194 | if (cdf_n < 0.0) #> | #> stan/lib/stan_math/stan/math/prim/prob/von_mises_cdf.hpp:194: note: adding '-flarge-source-files' will allow for more column-tracking support, at the expense of compilation time and memory -#> Start sampling -#> Running MCMC with 2 parallel chains... -#> -#> Chain 1 Iteration: 1 / 1000 [ 0%] (Warmup) -#> Chain 2 Iteration: 1 / 1000 [ 0%] (Warmup) -#> Chain 1 Iteration: 100 / 1000 [ 10%] (Warmup) -#> Chain 1 Iteration: 200 / 1000 [ 20%] (Warmup) -#> Chain 1 Iteration: 300 / 1000 [ 30%] (Warmup) -#> Chain 1 Iteration: 400 / 1000 [ 40%] (Warmup) -#> Chain 2 Iteration: 100 / 1000 [ 10%] (Warmup) -#> Chain 1 Iteration: 500 / 1000 [ 50%] (Warmup) -#> Chain 1 Iteration: 501 / 1000 [ 50%] (Sampling) -#> Chain 2 Iteration: 200 / 1000 [ 20%] (Warmup) -#> Chain 2 Iteration: 300 / 1000 [ 30%] (Warmup) -#> Chain 1 Iteration: 600 / 1000 [ 60%] (Sampling) -#> Chain 2 Iteration: 400 / 1000 [ 40%] (Warmup) -#> Chain 1 Iteration: 700 / 1000 [ 70%] (Sampling) -#> Chain 2 Iteration: 500 / 1000 [ 50%] (Warmup) -#> Chain 2 Iteration: 501 / 1000 [ 50%] (Sampling) -#> Chain 1 Iteration: 800 / 1000 [ 80%] (Sampling) -#> Chain 2 Iteration: 600 / 1000 [ 60%] (Sampling) -#> Chain 1 Iteration: 900 / 1000 [ 90%] (Sampling) -#> Chain 2 Iteration: 700 / 1000 [ 70%] (Sampling) -#> Chain 1 Iteration: 1000 / 1000 [100%] (Sampling) -#> Chain 2 Iteration: 800 / 1000 [ 80%] (Sampling) -#> Chain 1 finished in 1.0 seconds. -#> Chain 2 Iteration: 900 / 1000 [ 90%] (Sampling) -#> Chain 2 Iteration: 1000 / 1000 [100%] (Sampling) -#> Chain 2 finished in 1.1 seconds. -#> -#> Both chains finished successfully. -#> Mean chain execution time: 1.0 seconds. -#> Total execution time: 1.3 seconds. -#> # Hindcasts on response scale hc <- hindcast(mod) str(hc) #> List of 15 #> $ call :Class 'formula' language y ~ s(season, bs = "cc", k = 6) -#> .. ..- attr(*, ".Environment")=<environment: 0x0000022097b26560> +#> .. ..- attr(*, ".Environment")=<environment: 0x0000024633285b90> #> $ trend_call : NULL #> $ family : chr "poisson" #> $ trend_model :List of 7 @@ -215,22 +180,22 @@

Examples#> $ type : chr "response" #> $ series_names : chr [1:3] "series_1" "series_2" "series_3" #> $ train_observations:List of 3 -#> ..$ series_1: int [1:75] 0 0 0 1 0 1 1 4 0 0 ... -#> ..$ series_2: int [1:75] 0 0 0 0 1 0 4 0 0 0 ... -#> ..$ series_3: int [1:75] 1 1 1 2 0 1 4 6 6 0 ... +#> ..$ series_1: int [1:75] 0 1 0 0 1 2 0 3 0 0 ... +#> ..$ series_2: int [1:75] 0 4 0 0 0 2 1 0 0 1 ... +#> ..$ series_3: int [1:75] 2 3 3 1 1 8 4 5 0 0 ... #> $ train_times : int [1:75] 1 2 3 4 5 6 7 8 9 10 ... #> $ test_observations : NULL #> $ test_times : NULL #> $ hindcasts :List of 3 -#> ..$ series_1: num [1:1000, 1:75] 0 0 0 1 0 1 0 0 4 0 ... +#> ..$ series_1: num [1:1000, 1:75] 0 0 0 0 1 0 1 0 1 0 ... #> .. ..- attr(*, "dimnames")=List of 2 #> .. .. ..$ : NULL #> .. .. ..$ : chr [1:75] "ypred[1,1]" "ypred[2,1]" "ypred[3,1]" "ypred[4,1]" ... -#> ..$ series_2: num [1:1000, 1:75] 0 0 0 0 1 1 1 1 1 1 ... +#> ..$ series_2: num [1:1000, 1:75] 0 2 1 0 2 1 0 0 0 0 ... #> .. ..- attr(*, "dimnames")=List of 2 #> .. .. ..$ : NULL #> .. .. ..$ : chr [1:75] "ypred[1,2]" "ypred[2,2]" "ypred[3,2]" "ypred[4,2]" ... -#> ..$ series_3: num [1:1000, 1:75] 3 0 1 0 1 1 2 1 1 1 ... +#> ..$ series_3: num [1:1000, 1:75] 1 0 3 0 1 1 1 1 0 0 ... #> .. ..- attr(*, "dimnames")=List of 2 #> .. .. ..$ : NULL #> .. .. ..$ : chr [1:75] "ypred[1,3]" "ypred[2,3]" "ypred[3,3]" "ypred[4,3]" ... @@ -251,7 +216,7 @@

Examplesstr(fc) #> List of 16 #> $ call :Class 'formula' language y ~ s(season, bs = "cc", k = 6) -#> .. ..- attr(*, ".Environment")=<environment: 0x0000022097b26560> +#> .. ..- attr(*, ".Environment")=<environment: 0x0000024633285b90> #> $ trend_call : NULL #> $ family : chr "poisson" #> $ family_pars : NULL @@ -270,44 +235,44 @@

Examples#> $ type : chr "response" #> $ series_names : Factor w/ 3 levels "series_1","series_2",..: 1 2 3 #> $ train_observations:List of 3 -#> ..$ series_1: int [1:75] 0 0 0 1 0 1 1 4 0 0 ... -#> ..$ series_2: int [1:75] 0 0 0 0 1 0 4 0 0 0 ... -#> ..$ series_3: int [1:75] 1 1 1 2 0 1 4 6 6 0 ... +#> ..$ series_1: int [1:75] 0 1 0 0 1 2 0 3 0 0 ... +#> ..$ series_2: int [1:75] 0 4 0 0 0 2 1 0 0 1 ... +#> ..$ series_3: int [1:75] 2 3 3 1 1 8 4 5 0 0 ... #> $ train_times : int [1:75] 1 2 3 4 5 6 7 8 9 10 ... #> $ test_observations :List of 3 -#> ..$ series_1: int [1:25] 2 0 1 0 0 1 1 1 0 0 ... -#> ..$ series_2: int [1:25] 4 0 0 1 2 1 2 1 2 0 ... -#> ..$ series_3: int [1:25] 5 0 2 6 6 4 1 8 2 0 ... +#> ..$ series_1: int [1:25] 2 0 0 3 3 0 0 0 0 0 ... +#> ..$ series_2: int [1:25] 1 4 3 0 1 0 0 0 0 0 ... +#> ..$ series_3: int [1:25] 4 2 2 1 7 4 1 0 1 0 ... #> $ test_times : int [1:25] 76 77 78 79 80 81 82 83 84 85 ... #> $ hindcasts :List of 3 -#> ..$ series_1: num [1:1000, 1:75] 0 0 0 1 0 1 0 0 4 0 ... +#> ..$ series_1: num [1:1000, 1:75] 0 0 0 0 1 0 1 0 1 0 ... #> .. ..- attr(*, "dimnames")=List of 2 #> .. .. ..$ : NULL #> .. .. ..$ : chr [1:75] "ypred[1,1]" "ypred[2,1]" "ypred[3,1]" "ypred[4,1]" ... -#> ..$ series_2: num [1:1000, 1:75] 0 0 0 0 1 1 1 1 1 1 ... +#> ..$ series_2: num [1:1000, 1:75] 0 2 1 0 2 1 0 0 0 0 ... #> .. ..- attr(*, "dimnames")=List of 2 #> .. .. ..$ : NULL #> .. .. ..$ : chr [1:75] "ypred[1,2]" "ypred[2,2]" "ypred[3,2]" "ypred[4,2]" ... -#> ..$ series_3: num [1:1000, 1:75] 3 0 1 0 1 1 2 1 1 1 ... +#> ..$ series_3: num [1:1000, 1:75] 1 0 3 0 1 1 1 1 0 0 ... #> .. ..- attr(*, "dimnames")=List of 2 #> .. .. ..$ : NULL #> .. .. ..$ : chr [1:75] "ypred[1,3]" "ypred[2,3]" "ypred[3,3]" "ypred[4,3]" ... #> $ forecasts :List of 3 -#> ..$ series_1: int [1:1000, 1:25] 0 0 0 0 0 0 1 1 1 0 ... -#> ..$ series_2: int [1:1000, 1:25] 0 1 0 0 2 2 0 0 1 1 ... -#> ..$ series_3: int [1:1000, 1:25] 3 0 0 0 3 1 0 0 0 0 ... +#> ..$ series_1: int [1:1000, 1:25] 1 1 0 1 1 1 1 3 0 1 ... +#> ..$ series_2: int [1:1000, 1:25] 1 0 3 0 1 4 1 2 1 2 ... +#> ..$ series_3: int [1:1000, 1:25] 0 1 3 5 4 1 6 6 5 0 ... #> - attr(*, "class")= chr "mvgam_forecast" plot(fc, series = 1) #> Out of sample DRPS: -#> 10.868257 +#> 11.14604 plot(fc, series = 2) #> Out of sample DRPS: -#> 12.139955 +#> 14.552014 plot(fc, series = 3) #> Out of sample DRPS: -#> 32.02099 +#> 20.031465 # Forecasts as expectations diff --git a/docs/reference/lfo_cv.mvgam.html b/docs/reference/lfo_cv.mvgam.html index 4d5a6f59..38a09d07 100644 --- a/docs/reference/lfo_cv.mvgam.html +++ b/docs/reference/lfo_cv.mvgam.html @@ -1,5 +1,5 @@ -Approximate leave-future-out cross-validation of fitted mvgam objects — lfo_cv.mvgam • mvgamApproximate leave-future-out cross-validation of fitted mvgam objects — lfo_cv.mvgam • mvgam @@ -52,7 +52,7 @@
-

Approximate leave-future-out cross-validation of fitted mvgam objects

+

Approximate leave-future-out cross-validation of fitted mvgam objects

Source: R/lfo_cv.mvgam.R
lfo_cv.mvgam.Rd
@@ -191,7 +191,8 @@

Examples newdata = dat$data_test, burnin = 300, samples = 300, - chains = 2) + chains = 2, + silent = 2) # Fit a less appropriate model mod_rw <- mvgam(y ~ s(season, bs = 'cc', k = 6), @@ -201,7 +202,8 @@

Examples newdata = dat$data_test, burnin = 300, samples = 300, - chains = 2) + chains = 2, + silent = 2) # Compare Discrete Ranked Probability Scores for the testing period fc_ar2 <- forecast(mod_ar2) @@ -217,10 +219,12 @@

Examples# for estimating model parameters lfo_ar2 <- lfo_cv(mod_ar2, min_t = 40, - fc_horizon = 3) + fc_horizon = 3, + silent = 2) lfo_rw <- lfo_cv(mod_rw, min_t = 40, - fc_horizon = 3) + fc_horizon = 3, + silent = 2) # Plot Pareto-K values and ELPD estimates plot(lfo_ar2) diff --git a/docs/reference/loo.mvgam-1.png b/docs/reference/loo.mvgam-1.png index 41070110..2c2aff89 100644 Binary files a/docs/reference/loo.mvgam-1.png and b/docs/reference/loo.mvgam-1.png differ diff --git a/docs/reference/loo.mvgam-2.png b/docs/reference/loo.mvgam-2.png index ae40aeba..17599ad9 100644 Binary files a/docs/reference/loo.mvgam-2.png and b/docs/reference/loo.mvgam-2.png differ diff --git a/docs/reference/loo.mvgam-3.png b/docs/reference/loo.mvgam-3.png index 7939f149..ec9d50e3 100644 Binary files a/docs/reference/loo.mvgam-3.png and b/docs/reference/loo.mvgam-3.png differ diff --git a/docs/reference/loo.mvgam-4.png b/docs/reference/loo.mvgam-4.png index 719bce1a..706d040d 100644 Binary files a/docs/reference/loo.mvgam-4.png and b/docs/reference/loo.mvgam-4.png differ diff --git a/docs/reference/loo.mvgam-5.png b/docs/reference/loo.mvgam-5.png index 7efd656e..6a45504e 100644 Binary files a/docs/reference/loo.mvgam-5.png and b/docs/reference/loo.mvgam-5.png differ diff --git a/docs/reference/loo.mvgam-6.png b/docs/reference/loo.mvgam-6.png index 52c735d8..adcbeb61 100644 Binary files a/docs/reference/loo.mvgam-6.png and b/docs/reference/loo.mvgam-6.png differ diff --git a/docs/reference/loo.mvgam-7.png b/docs/reference/loo.mvgam-7.png index e1c28134..6c4bc345 100644 Binary files a/docs/reference/loo.mvgam-7.png and b/docs/reference/loo.mvgam-7.png differ diff --git a/docs/reference/loo.mvgam.html b/docs/reference/loo.mvgam.html index 053e8ed0..bc9ab038 100644 --- a/docs/reference/loo.mvgam.html +++ b/docs/reference/loo.mvgam.html @@ -148,43 +148,21 @@

Examples data = rbind(simdat$data_train, simdat$data_test), family = gaussian(), - chains = 2) -#> Compiling Stan program using cmdstanr -#> -#> Start sampling -#> Running MCMC with 2 parallel chains... -#> -#> Chain 1 Iteration: 1 / 1000 [ 0%] (Warmup) -#> Chain 1 Iteration: 100 / 1000 [ 10%] (Warmup) -#> Chain 1 Iteration: 200 / 1000 [ 20%] (Warmup) -#> Chain 1 Iteration: 300 / 1000 [ 30%] (Warmup) -#> Chain 1 Iteration: 400 / 1000 [ 40%] (Warmup) -#> Chain 1 Iteration: 500 / 1000 [ 50%] (Warmup) -#> Chain 1 Iteration: 501 / 1000 [ 50%] (Sampling) -#> Chain 1 Iteration: 600 / 1000 [ 60%] (Sampling) -#> Chain 2 Iteration: 1 / 1000 [ 0%] (Warmup) -#> Chain 2 Iteration: 100 / 1000 [ 10%] (Warmup) -#> Chain 2 Iteration: 200 / 1000 [ 20%] (Warmup) -#> Chain 2 Iteration: 300 / 1000 [ 30%] (Warmup) -#> Chain 2 Iteration: 400 / 1000 [ 40%] (Warmup) -#> Chain 2 Iteration: 500 / 1000 [ 50%] (Warmup) -#> Chain 2 Iteration: 501 / 1000 [ 50%] (Sampling) -#> Chain 1 Iteration: 700 / 1000 [ 70%] (Sampling) -#> Chain 2 Iteration: 600 / 1000 [ 60%] (Sampling) -#> Chain 1 Iteration: 800 / 1000 [ 80%] (Sampling) -#> Chain 2 Iteration: 700 / 1000 [ 70%] (Sampling) -#> Chain 1 Iteration: 900 / 1000 [ 90%] (Sampling) -#> Chain 2 Iteration: 800 / 1000 [ 80%] (Sampling) -#> Chain 1 Iteration: 1000 / 1000 [100%] (Sampling) -#> Chain 2 Iteration: 900 / 1000 [ 90%] (Sampling) -#> Chain 1 finished in 0.7 seconds. -#> Chain 2 Iteration: 1000 / 1000 [100%] (Sampling) -#> Chain 2 finished in 0.7 seconds. -#> -#> Both chains finished successfully. -#> Mean chain execution time: 0.7 seconds. -#> Total execution time: 1.1 seconds. -#> + chains = 2, + silent = 2) +#> In file included from stan/lib/stan_math/stan/math/prim/prob/von_mises_lccdf.hpp:5, +#> from stan/lib/stan_math/stan/math/prim/prob/von_mises_ccdf_log.hpp:4, +#> from stan/lib/stan_math/stan/math/prim/prob.hpp:359, +#> from stan/lib/stan_math/stan/math/prim.hpp:16, +#> from stan/lib/stan_math/stan/math/rev.hpp:16, +#> from stan/lib/stan_math/stan/math.hpp:19, +#> from stan/src/stan/model/model_header.hpp:4, +#> from C:/Users/uqnclar2/AppData/Local/Temp/RtmpqqsueL/model-19b8407b5d02.hpp:2: +#> stan/lib/stan_math/stan/math/prim/prob/von_mises_cdf.hpp: In function 'stan::return_type_t<T_x, T_sigma, T_l> stan::math::von_mises_cdf(const T_x&, const T_mu&, const T_k&)': +#> stan/lib/stan_math/stan/math/prim/prob/von_mises_cdf.hpp:194: note: '-Wmisleading-indentation' is disabled from this point onwards, since column-tracking was disabled due to the size of the code/headers +#> 194 | if (cdf_n < 0.0) +#> | +#> stan/lib/stan_math/stan/math/prim/prob/von_mises_cdf.hpp:194: note: adding '-flarge-source-files' will allow for more column-tracking support, at the expense of compilation time and memory # Inspect the model and calculate LOO conditional_effects(mod1) @@ -196,12 +174,12 @@

Examples#> Computed from 1000 by 300 log-likelihood matrix. #> #> Estimate SE -#> elpd_loo -363.8 11.1 -#> p_loo 7.0 0.6 -#> looic 727.6 22.1 +#> elpd_loo -364.2 11.1 +#> p_loo 7.4 0.6 +#> looic 728.4 22.1 #> ------ #> MCSE of elpd_loo is 0.1. -#> MCSE and ESS estimates assume MCMC draws (r_eff in [0.9, 1.9]). +#> MCSE and ESS estimates assume MCMC draws (r_eff in [0.7, 2.3]). #> #> All Pareto k estimates are good (k < 0.67). #> See help('pareto-k-diagnostic') for details. @@ -211,46 +189,24 @@

Examples formula = y ~ s(season, bs = 'cc', k = 6) + s(season, series, bs = 'fs', xt = list(bs = 'cc'), k = 4), - chains = 2) + chains = 2, + silent = 2) #> Warning: model has repeated 1-d smooths of same variable. #> Warning: model has repeated 1-d smooths of same variable. #> Warning: model has repeated 1-d smooths of same variable. -#> Compiling Stan program using cmdstanr -#> -#> Start sampling -#> Running MCMC with 2 parallel chains... -#> -#> Chain 1 Iteration: 1 / 1000 [ 0%] (Warmup) -#> Chain 1 Iteration: 100 / 1000 [ 10%] (Warmup) -#> Chain 2 Iteration: 1 / 1000 [ 0%] (Warmup) -#> Chain 1 Iteration: 200 / 1000 [ 20%] (Warmup) -#> Chain 2 Iteration: 100 / 1000 [ 10%] (Warmup) -#> Chain 2 Iteration: 200 / 1000 [ 20%] (Warmup) -#> Chain 1 Iteration: 300 / 1000 [ 30%] (Warmup) -#> Chain 2 Iteration: 300 / 1000 [ 30%] (Warmup) -#> Chain 2 Iteration: 400 / 1000 [ 40%] (Warmup) -#> Chain 2 Iteration: 500 / 1000 [ 50%] (Warmup) -#> Chain 2 Iteration: 501 / 1000 [ 50%] (Sampling) -#> Chain 1 Iteration: 400 / 1000 [ 40%] (Warmup) -#> Chain 1 Iteration: 500 / 1000 [ 50%] (Warmup) -#> Chain 2 Iteration: 600 / 1000 [ 60%] (Sampling) -#> Chain 1 Iteration: 501 / 1000 [ 50%] (Sampling) -#> Chain 1 Iteration: 600 / 1000 [ 60%] (Sampling) -#> Chain 2 Iteration: 700 / 1000 [ 70%] (Sampling) -#> Chain 1 Iteration: 700 / 1000 [ 70%] (Sampling) -#> Chain 2 Iteration: 800 / 1000 [ 80%] (Sampling) -#> Chain 1 Iteration: 800 / 1000 [ 80%] (Sampling) -#> Chain 2 Iteration: 900 / 1000 [ 90%] (Sampling) -#> Chain 1 Iteration: 900 / 1000 [ 90%] (Sampling) -#> Chain 2 Iteration: 1000 / 1000 [100%] (Sampling) -#> Chain 2 finished in 2.3 seconds. -#> Chain 1 Iteration: 1000 / 1000 [100%] (Sampling) -#> Chain 1 finished in 2.6 seconds. -#> -#> Both chains finished successfully. -#> Mean chain execution time: 2.5 seconds. -#> Total execution time: 2.8 seconds. -#> +#> In file included from stan/lib/stan_math/stan/math/prim/prob/von_mises_lccdf.hpp:5, +#> from stan/lib/stan_math/stan/math/prim/prob/von_mises_ccdf_log.hpp:4, +#> from stan/lib/stan_math/stan/math/prim/prob.hpp:359, +#> from stan/lib/stan_math/stan/math/prim.hpp:16, +#> from stan/lib/stan_math/stan/math/rev.hpp:16, +#> from stan/lib/stan_math/stan/math.hpp:19, +#> from stan/src/stan/model/model_header.hpp:4, +#> from C:/Users/uqnclar2/AppData/Local/Temp/RtmpqqsueL/model-19b830efb18.hpp:2: +#> stan/lib/stan_math/stan/math/prim/prob/von_mises_cdf.hpp: In function 'stan::return_type_t<T_x, T_sigma, T_l> stan::math::von_mises_cdf(const T_x&, const T_mu&, const T_k&)': +#> stan/lib/stan_math/stan/math/prim/prob/von_mises_cdf.hpp:194: note: '-Wmisleading-indentation' is disabled from this point onwards, since column-tracking was disabled due to the size of the code/headers +#> 194 | if (cdf_n < 0.0) +#> | +#> stan/lib/stan_math/stan/math/prim/prob/von_mises_cdf.hpp:194: note: adding '-flarge-source-files' will allow for more column-tracking support, at the expense of compilation time and memory conditional_effects(mod2) @@ -259,12 +215,12 @@

Examples#> Computed from 1000 by 300 log-likelihood matrix. #> #> Estimate SE -#> elpd_loo -309.4 11.4 -#> p_loo 12.9 1.0 -#> looic 618.8 22.7 +#> elpd_loo -309.7 11.4 +#> p_loo 13.3 1.1 +#> looic 619.5 22.8 #> ------ #> MCSE of elpd_loo is 0.1. -#> MCSE and ESS estimates assume MCMC draws (r_eff in [0.6, 1.2]). +#> MCSE and ESS estimates assume MCMC draws (r_eff in [0.7, 1.3]). #> #> All Pareto k estimates are good (k < 0.67). #> See help('pareto-k-diagnostic') for details. @@ -272,47 +228,24 @@

Examples# Now add AR1 dynamic errors to mod2 mod3 <- update(mod2, trend_model = AR(), - chains = 2) + chains = 2, + silent = 2) #> Warning: model has repeated 1-d smooths of same variable. #> Warning: model has repeated 1-d smooths of same variable. #> Warning: model has repeated 1-d smooths of same variable. -#> Your model may benefit from using "noncentred = TRUE" -#> Compiling Stan program using cmdstanr -#> -#> Start sampling -#> Running MCMC with 2 parallel chains... -#> -#> Chain 1 Iteration: 1 / 1000 [ 0%] (Warmup) -#> Chain 2 Iteration: 1 / 1000 [ 0%] (Warmup) -#> Chain 1 Iteration: 100 / 1000 [ 10%] (Warmup) -#> Chain 2 Iteration: 100 / 1000 [ 10%] (Warmup) -#> Chain 1 Iteration: 200 / 1000 [ 20%] (Warmup) -#> Chain 2 Iteration: 200 / 1000 [ 20%] (Warmup) -#> Chain 2 Iteration: 300 / 1000 [ 30%] (Warmup) -#> Chain 1 Iteration: 300 / 1000 [ 30%] (Warmup) -#> Chain 2 Iteration: 400 / 1000 [ 40%] (Warmup) -#> Chain 2 Iteration: 500 / 1000 [ 50%] (Warmup) -#> Chain 1 Iteration: 400 / 1000 [ 40%] (Warmup) -#> Chain 2 Iteration: 501 / 1000 [ 50%] (Sampling) -#> Chain 2 Iteration: 600 / 1000 [ 60%] (Sampling) -#> Chain 1 Iteration: 500 / 1000 [ 50%] (Warmup) -#> Chain 1 Iteration: 501 / 1000 [ 50%] (Sampling) -#> Chain 2 Iteration: 700 / 1000 [ 70%] (Sampling) -#> Chain 2 Iteration: 800 / 1000 [ 80%] (Sampling) -#> Chain 1 Iteration: 600 / 1000 [ 60%] (Sampling) -#> Chain 2 Iteration: 900 / 1000 [ 90%] (Sampling) -#> Chain 1 Iteration: 700 / 1000 [ 70%] (Sampling) -#> Chain 2 Iteration: 1000 / 1000 [100%] (Sampling) -#> Chain 2 finished in 10.1 seconds. -#> Chain 1 Iteration: 800 / 1000 [ 80%] (Sampling) -#> Chain 1 Iteration: 900 / 1000 [ 90%] (Sampling) -#> Chain 1 Iteration: 1000 / 1000 [100%] (Sampling) -#> Chain 1 finished in 14.6 seconds. -#> -#> Both chains finished successfully. -#> Mean chain execution time: 12.4 seconds. -#> Total execution time: 14.8 seconds. -#> +#> In file included from stan/lib/stan_math/stan/math/prim/prob/von_mises_lccdf.hpp:5, +#> from stan/lib/stan_math/stan/math/prim/prob/von_mises_ccdf_log.hpp:4, +#> from stan/lib/stan_math/stan/math/prim/prob.hpp:359, +#> from stan/lib/stan_math/stan/math/prim.hpp:16, +#> from stan/lib/stan_math/stan/math/rev.hpp:16, +#> from stan/lib/stan_math/stan/math.hpp:19, +#> from stan/src/stan/model/model_header.hpp:4, +#> from C:/Users/uqnclar2/AppData/Local/Temp/RtmpqqsueL/model-19b84ce92891.hpp:2: +#> stan/lib/stan_math/stan/math/prim/prob/von_mises_cdf.hpp: In function 'stan::return_type_t<T_x, T_sigma, T_l> stan::math::von_mises_cdf(const T_x&, const T_mu&, const T_k&)': +#> stan/lib/stan_math/stan/math/prim/prob/von_mises_cdf.hpp:194: note: '-Wmisleading-indentation' is disabled from this point onwards, since column-tracking was disabled due to the size of the code/headers +#> 194 | if (cdf_n < 0.0) +#> | +#> stan/lib/stan_math/stan/math/prim/prob/von_mises_cdf.hpp:194: note: adding '-flarge-source-files' will allow for more column-tracking support, at the expense of compilation time and memory conditional_effects(mod3) @@ -324,18 +257,18 @@

Examples#> Computed from 1000 by 300 log-likelihood matrix. #> #> Estimate SE -#> elpd_loo -261.8 10.1 -#> p_loo 155.4 7.0 -#> looic 523.7 20.3 +#> elpd_loo -251.0 9.9 +#> p_loo 163.6 7.1 +#> looic 502.0 19.9 #> ------ #> MCSE of elpd_loo is NA. -#> MCSE and ESS estimates assume MCMC draws (r_eff in [0.0, 0.1]). +#> MCSE and ESS estimates assume MCMC draws (r_eff in [0.0, 0.0]). #> #> Pareto k diagnostic values: #> Count Pct. Min. ESS -#> (-Inf, 0.67] (good) 151 50.3% 1 -#> (0.67, 1] (bad) 142 47.3% <NA> -#> (1, Inf) (very bad) 7 2.3% <NA> +#> (-Inf, 0.67] (good) 127 42.3% 1 +#> (0.67, 1] (bad) 160 53.3% <NA> +#> (1, Inf) (very bad) 13 4.3% <NA> #> See help('pareto-k-diagnostic') for details. # Compare models using LOO @@ -343,8 +276,8 @@

Examples#> Warning: Some Pareto k diagnostic values are too high. See help('pareto-k-diagnostic') for details. #> elpd_diff se_diff #> mod3 0.0 0.0 -#> mod2 -47.6 4.3 -#> mod1 -101.9 9.1 +#> mod2 -58.7 4.8 +#> mod1 -113.2 9.3 options(mc.cores = mc.cores.def) # Compare forecast abilities using an expanding training window and @@ -358,39 +291,52 @@

Examples#> Warning: model has repeated 1-d smooths of same variable. #> Compiling Stan program using cmdstanr #> +#> In file included from stan/lib/stan_math/stan/math/prim/prob/von_mises_lccdf.hpp:5, +#> from stan/lib/stan_math/stan/math/prim/prob/von_mises_ccdf_log.hpp:4, +#> from stan/lib/stan_math/stan/math/prim/prob.hpp:359, +#> from stan/lib/stan_math/stan/math/prim.hpp:16, +#> from stan/lib/stan_math/stan/math/rev.hpp:16, +#> from stan/lib/stan_math/stan/math.hpp:19, +#> from stan/src/stan/model/model_header.hpp:4, +#> from C:/Users/uqnclar2/AppData/Local/Temp/RtmpqqsueL/model-19b824896df3.hpp:2: +#> stan/lib/stan_math/stan/math/prim/prob/von_mises_cdf.hpp: In function 'stan::return_type_t<T_x, T_sigma, T_l> stan::math::von_mises_cdf(const T_x&, const T_mu&, const T_k&)': +#> stan/lib/stan_math/stan/math/prim/prob/von_mises_cdf.hpp:194: note: '-Wmisleading-indentation' is disabled from this point onwards, since column-tracking was disabled due to the size of the code/headers +#> 194 | if (cdf_n < 0.0) +#> | +#> stan/lib/stan_math/stan/math/prim/prob/von_mises_cdf.hpp:194: note: adding '-flarge-source-files' will allow for more column-tracking support, at the expense of compilation time and memory #> Start sampling #> Running MCMC with 2 parallel chains... #> #> Chain 1 Iteration: 1 / 1000 [ 0%] (Warmup) #> Chain 1 Iteration: 100 / 1000 [ 10%] (Warmup) +#> Chain 1 Iteration: 200 / 1000 [ 20%] (Warmup) #> Chain 2 Iteration: 1 / 1000 [ 0%] (Warmup) #> Chain 2 Iteration: 100 / 1000 [ 10%] (Warmup) -#> Chain 1 Iteration: 200 / 1000 [ 20%] (Warmup) -#> Chain 2 Iteration: 200 / 1000 [ 20%] (Warmup) #> Chain 1 Iteration: 300 / 1000 [ 30%] (Warmup) -#> Chain 2 Iteration: 300 / 1000 [ 30%] (Warmup) #> Chain 1 Iteration: 400 / 1000 [ 40%] (Warmup) -#> Chain 2 Iteration: 400 / 1000 [ 40%] (Warmup) #> Chain 1 Iteration: 500 / 1000 [ 50%] (Warmup) -#> Chain 2 Iteration: 500 / 1000 [ 50%] (Warmup) #> Chain 1 Iteration: 501 / 1000 [ 50%] (Sampling) -#> Chain 2 Iteration: 501 / 1000 [ 50%] (Sampling) +#> Chain 2 Iteration: 200 / 1000 [ 20%] (Warmup) +#> Chain 2 Iteration: 300 / 1000 [ 30%] (Warmup) +#> Chain 2 Iteration: 400 / 1000 [ 40%] (Warmup) #> Chain 1 Iteration: 600 / 1000 [ 60%] (Sampling) -#> Chain 2 Iteration: 600 / 1000 [ 60%] (Sampling) +#> Chain 2 Iteration: 500 / 1000 [ 50%] (Warmup) +#> Chain 2 Iteration: 501 / 1000 [ 50%] (Sampling) #> Chain 1 Iteration: 700 / 1000 [ 70%] (Sampling) +#> Chain 2 Iteration: 600 / 1000 [ 60%] (Sampling) +#> Chain 1 Iteration: 800 / 1000 [ 80%] (Sampling) #> Chain 2 Iteration: 700 / 1000 [ 70%] (Sampling) +#> Chain 1 Iteration: 900 / 1000 [ 90%] (Sampling) #> Chain 2 Iteration: 800 / 1000 [ 80%] (Sampling) -#> Chain 1 Iteration: 800 / 1000 [ 80%] (Sampling) +#> Chain 1 Iteration: 1000 / 1000 [100%] (Sampling) #> Chain 2 Iteration: 900 / 1000 [ 90%] (Sampling) -#> Chain 1 Iteration: 900 / 1000 [ 90%] (Sampling) +#> Chain 1 finished in 1.0 seconds. #> Chain 2 Iteration: 1000 / 1000 [100%] (Sampling) -#> Chain 2 finished in 1.9 seconds. -#> Chain 1 Iteration: 1000 / 1000 [100%] (Sampling) -#> Chain 1 finished in 2.2 seconds. +#> Chain 2 finished in 1.2 seconds. #> #> Both chains finished successfully. -#> Mean chain execution time: 2.0 seconds. -#> Total execution time: 2.4 seconds. +#> Mean chain execution time: 1.1 seconds. +#> Total execution time: 1.3 seconds. #> lfo_mod3 <- lfo_cv(mod3, min_t = 92) #> Warning: model has repeated 1-d smooths of same variable. @@ -399,39 +345,52 @@

Examples#> Your model may benefit from using "noncentred = TRUE" #> Compiling Stan program using cmdstanr #> +#> In file included from stan/lib/stan_math/stan/math/prim/prob/von_mises_lccdf.hpp:5, +#> from stan/lib/stan_math/stan/math/prim/prob/von_mises_ccdf_log.hpp:4, +#> from stan/lib/stan_math/stan/math/prim/prob.hpp:359, +#> from stan/lib/stan_math/stan/math/prim.hpp:16, +#> from stan/lib/stan_math/stan/math/rev.hpp:16, +#> from stan/lib/stan_math/stan/math.hpp:19, +#> from stan/src/stan/model/model_header.hpp:4, +#> from C:/Users/uqnclar2/AppData/Local/Temp/RtmpqqsueL/model-19b85d765643.hpp:2: +#> stan/lib/stan_math/stan/math/prim/prob/von_mises_cdf.hpp: In function 'stan::return_type_t<T_x, T_sigma, T_l> stan::math::von_mises_cdf(const T_x&, const T_mu&, const T_k&)': +#> stan/lib/stan_math/stan/math/prim/prob/von_mises_cdf.hpp:194: note: '-Wmisleading-indentation' is disabled from this point onwards, since column-tracking was disabled due to the size of the code/headers +#> 194 | if (cdf_n < 0.0) +#> | +#> stan/lib/stan_math/stan/math/prim/prob/von_mises_cdf.hpp:194: note: adding '-flarge-source-files' will allow for more column-tracking support, at the expense of compilation time and memory #> Start sampling #> Running MCMC with 2 parallel chains... #> #> Chain 1 Iteration: 1 / 1000 [ 0%] (Warmup) #> Chain 2 Iteration: 1 / 1000 [ 0%] (Warmup) -#> Chain 1 Iteration: 100 / 1000 [ 10%] (Warmup) #> Chain 2 Iteration: 100 / 1000 [ 10%] (Warmup) +#> Chain 1 Iteration: 100 / 1000 [ 10%] (Warmup) #> Chain 2 Iteration: 200 / 1000 [ 20%] (Warmup) #> Chain 1 Iteration: 200 / 1000 [ 20%] (Warmup) -#> Chain 1 Iteration: 300 / 1000 [ 30%] (Warmup) #> Chain 2 Iteration: 300 / 1000 [ 30%] (Warmup) -#> Chain 1 Iteration: 400 / 1000 [ 40%] (Warmup) #> Chain 2 Iteration: 400 / 1000 [ 40%] (Warmup) +#> Chain 1 Iteration: 300 / 1000 [ 30%] (Warmup) #> Chain 2 Iteration: 500 / 1000 [ 50%] (Warmup) #> Chain 2 Iteration: 501 / 1000 [ 50%] (Sampling) -#> Chain 1 Iteration: 500 / 1000 [ 50%] (Warmup) -#> Chain 1 Iteration: 501 / 1000 [ 50%] (Sampling) #> Chain 2 Iteration: 600 / 1000 [ 60%] (Sampling) +#> Chain 1 Iteration: 400 / 1000 [ 40%] (Warmup) #> Chain 2 Iteration: 700 / 1000 [ 70%] (Sampling) #> Chain 2 Iteration: 800 / 1000 [ 80%] (Sampling) +#> Chain 1 Iteration: 500 / 1000 [ 50%] (Warmup) #> Chain 2 Iteration: 900 / 1000 [ 90%] (Sampling) +#> Chain 1 Iteration: 501 / 1000 [ 50%] (Sampling) #> Chain 2 Iteration: 1000 / 1000 [100%] (Sampling) -#> Chain 2 finished in 12.8 seconds. +#> Chain 2 finished in 6.6 seconds. #> Chain 1 Iteration: 600 / 1000 [ 60%] (Sampling) #> Chain 1 Iteration: 700 / 1000 [ 70%] (Sampling) #> Chain 1 Iteration: 800 / 1000 [ 80%] (Sampling) #> Chain 1 Iteration: 900 / 1000 [ 90%] (Sampling) #> Chain 1 Iteration: 1000 / 1000 [100%] (Sampling) -#> Chain 1 finished in 36.4 seconds. +#> Chain 1 finished in 11.1 seconds. #> #> Both chains finished successfully. -#> Mean chain execution time: 24.6 seconds. -#> Total execution time: 36.7 seconds. +#> Mean chain execution time: 8.8 seconds. +#> Total execution time: 11.1 seconds. #> #> Warning: model has repeated 1-d smooths of same variable. #> Warning: model has repeated 1-d smooths of same variable. @@ -446,32 +405,32 @@

Examples#> Chain 2 Iteration: 1 / 1000 [ 0%] (Warmup) #> Chain 1 Iteration: 100 / 1000 [ 10%] (Warmup) #> Chain 2 Iteration: 100 / 1000 [ 10%] (Warmup) -#> Chain 1 Iteration: 200 / 1000 [ 20%] (Warmup) #> Chain 2 Iteration: 200 / 1000 [ 20%] (Warmup) +#> Chain 1 Iteration: 200 / 1000 [ 20%] (Warmup) #> Chain 1 Iteration: 300 / 1000 [ 30%] (Warmup) -#> Chain 1 Iteration: 400 / 1000 [ 40%] (Warmup) -#> Chain 1 Iteration: 500 / 1000 [ 50%] (Warmup) -#> Chain 1 Iteration: 501 / 1000 [ 50%] (Sampling) #> Chain 2 Iteration: 300 / 1000 [ 30%] (Warmup) -#> Chain 1 Iteration: 600 / 1000 [ 60%] (Sampling) #> Chain 2 Iteration: 400 / 1000 [ 40%] (Warmup) -#> Chain 1 Iteration: 700 / 1000 [ 70%] (Sampling) -#> Chain 1 Iteration: 800 / 1000 [ 80%] (Sampling) +#> Chain 1 Iteration: 400 / 1000 [ 40%] (Warmup) #> Chain 2 Iteration: 500 / 1000 [ 50%] (Warmup) +#> Chain 1 Iteration: 500 / 1000 [ 50%] (Warmup) #> Chain 2 Iteration: 501 / 1000 [ 50%] (Sampling) +#> Chain 1 Iteration: 501 / 1000 [ 50%] (Sampling) +#> Chain 1 Iteration: 600 / 1000 [ 60%] (Sampling) #> Chain 2 Iteration: 600 / 1000 [ 60%] (Sampling) -#> Chain 1 Iteration: 900 / 1000 [ 90%] (Sampling) +#> Chain 1 Iteration: 700 / 1000 [ 70%] (Sampling) #> Chain 2 Iteration: 700 / 1000 [ 70%] (Sampling) -#> Chain 1 Iteration: 1000 / 1000 [100%] (Sampling) -#> Chain 1 finished in 17.0 seconds. +#> Chain 1 Iteration: 800 / 1000 [ 80%] (Sampling) #> Chain 2 Iteration: 800 / 1000 [ 80%] (Sampling) #> Chain 2 Iteration: 900 / 1000 [ 90%] (Sampling) +#> Chain 1 Iteration: 900 / 1000 [ 90%] (Sampling) #> Chain 2 Iteration: 1000 / 1000 [100%] (Sampling) -#> Chain 2 finished in 21.1 seconds. +#> Chain 2 finished in 7.0 seconds. +#> Chain 1 Iteration: 1000 / 1000 [100%] (Sampling) +#> Chain 1 finished in 7.9 seconds. #> #> Both chains finished successfully. -#> Mean chain execution time: 19.1 seconds. -#> Total execution time: 21.5 seconds. +#> Mean chain execution time: 7.4 seconds. +#> Total execution time: 7.9 seconds. #> #> Warning: model has repeated 1-d smooths of same variable. #> Warning: model has repeated 1-d smooths of same variable. @@ -486,32 +445,32 @@

Examples#> Chain 2 Iteration: 1 / 1000 [ 0%] (Warmup) #> Chain 1 Iteration: 100 / 1000 [ 10%] (Warmup) #> Chain 2 Iteration: 100 / 1000 [ 10%] (Warmup) -#> Chain 2 Iteration: 200 / 1000 [ 20%] (Warmup) #> Chain 1 Iteration: 200 / 1000 [ 20%] (Warmup) +#> Chain 2 Iteration: 200 / 1000 [ 20%] (Warmup) #> Chain 1 Iteration: 300 / 1000 [ 30%] (Warmup) -#> Chain 2 Iteration: 300 / 1000 [ 30%] (Warmup) #> Chain 1 Iteration: 400 / 1000 [ 40%] (Warmup) -#> Chain 2 Iteration: 400 / 1000 [ 40%] (Warmup) #> Chain 1 Iteration: 500 / 1000 [ 50%] (Warmup) #> Chain 1 Iteration: 501 / 1000 [ 50%] (Sampling) +#> Chain 2 Iteration: 300 / 1000 [ 30%] (Warmup) +#> Chain 1 Iteration: 600 / 1000 [ 60%] (Sampling) +#> Chain 1 Iteration: 700 / 1000 [ 70%] (Sampling) +#> Chain 2 Iteration: 400 / 1000 [ 40%] (Warmup) +#> Chain 1 Iteration: 800 / 1000 [ 80%] (Sampling) +#> Chain 1 Iteration: 900 / 1000 [ 90%] (Sampling) +#> Chain 1 Iteration: 1000 / 1000 [100%] (Sampling) +#> Chain 1 finished in 8.8 seconds. #> Chain 2 Iteration: 500 / 1000 [ 50%] (Warmup) #> Chain 2 Iteration: 501 / 1000 [ 50%] (Sampling) -#> Chain 1 Iteration: 600 / 1000 [ 60%] (Sampling) #> Chain 2 Iteration: 600 / 1000 [ 60%] (Sampling) -#> Chain 1 Iteration: 700 / 1000 [ 70%] (Sampling) #> Chain 2 Iteration: 700 / 1000 [ 70%] (Sampling) #> Chain 2 Iteration: 800 / 1000 [ 80%] (Sampling) -#> Chain 1 Iteration: 800 / 1000 [ 80%] (Sampling) -#> Chain 1 Iteration: 900 / 1000 [ 90%] (Sampling) #> Chain 2 Iteration: 900 / 1000 [ 90%] (Sampling) #> Chain 2 Iteration: 1000 / 1000 [100%] (Sampling) -#> Chain 2 finished in 24.9 seconds. -#> Chain 1 Iteration: 1000 / 1000 [100%] (Sampling) -#> Chain 1 finished in 25.4 seconds. +#> Chain 2 finished in 25.0 seconds. #> #> Both chains finished successfully. -#> Mean chain execution time: 25.2 seconds. -#> Total execution time: 25.7 seconds. +#> Mean chain execution time: 16.9 seconds. +#> Total execution time: 25.2 seconds. #> # Take the difference in forecast ELPDs; a model with higher ELPD is preferred, diff --git a/docs/reference/mvgam-1.png b/docs/reference/mvgam-1.png index e328ce49..430cb723 100644 Binary files a/docs/reference/mvgam-1.png and b/docs/reference/mvgam-1.png differ diff --git a/docs/reference/mvgam-10.png b/docs/reference/mvgam-10.png index 65d6b581..2eaa0a02 100644 Binary files a/docs/reference/mvgam-10.png and b/docs/reference/mvgam-10.png differ diff --git a/docs/reference/mvgam-11.png b/docs/reference/mvgam-11.png index 61b0cff8..93b6a939 100644 Binary files a/docs/reference/mvgam-11.png and b/docs/reference/mvgam-11.png differ diff --git a/docs/reference/mvgam-12.png b/docs/reference/mvgam-12.png index 12c720a1..6153dd79 100644 Binary files a/docs/reference/mvgam-12.png and b/docs/reference/mvgam-12.png differ diff --git a/docs/reference/mvgam-13.png b/docs/reference/mvgam-13.png index 9db19eac..b18d849a 100644 Binary files a/docs/reference/mvgam-13.png and b/docs/reference/mvgam-13.png differ diff --git a/docs/reference/mvgam-14.png b/docs/reference/mvgam-14.png index 4fbfac7e..b7a6d24e 100644 Binary files a/docs/reference/mvgam-14.png and b/docs/reference/mvgam-14.png differ diff --git a/docs/reference/mvgam-15.png b/docs/reference/mvgam-15.png index 53c22aa9..274fc0c4 100644 Binary files a/docs/reference/mvgam-15.png and b/docs/reference/mvgam-15.png differ diff --git a/docs/reference/mvgam-16.png b/docs/reference/mvgam-16.png index 3eec9485..a580b628 100644 Binary files a/docs/reference/mvgam-16.png and b/docs/reference/mvgam-16.png differ diff --git a/docs/reference/mvgam-18.png b/docs/reference/mvgam-18.png index 3fe344b1..d57a4a5c 100644 Binary files a/docs/reference/mvgam-18.png and b/docs/reference/mvgam-18.png differ diff --git a/docs/reference/mvgam-19.png b/docs/reference/mvgam-19.png index 409f7c0d..0d9c672a 100644 Binary files a/docs/reference/mvgam-19.png and b/docs/reference/mvgam-19.png differ diff --git a/docs/reference/mvgam-2.png b/docs/reference/mvgam-2.png index b05f9ef4..8a176cf1 100644 Binary files a/docs/reference/mvgam-2.png and b/docs/reference/mvgam-2.png differ diff --git a/docs/reference/mvgam-20.png b/docs/reference/mvgam-20.png index 119662eb..cde55640 100644 Binary files a/docs/reference/mvgam-20.png and b/docs/reference/mvgam-20.png differ diff --git a/docs/reference/mvgam-21.png b/docs/reference/mvgam-21.png index c4f5049a..7f27ab77 100644 Binary files a/docs/reference/mvgam-21.png and b/docs/reference/mvgam-21.png differ diff --git a/docs/reference/mvgam-22.png b/docs/reference/mvgam-22.png index bb509d3a..0257cdd7 100644 Binary files a/docs/reference/mvgam-22.png and b/docs/reference/mvgam-22.png differ diff --git a/docs/reference/mvgam-23.png b/docs/reference/mvgam-23.png index 19fad5cb..c2c65dea 100644 Binary files a/docs/reference/mvgam-23.png and b/docs/reference/mvgam-23.png differ diff --git a/docs/reference/mvgam-24.png b/docs/reference/mvgam-24.png index b0639411..6c761a3a 100644 Binary files a/docs/reference/mvgam-24.png and b/docs/reference/mvgam-24.png differ diff --git a/docs/reference/mvgam-25.png b/docs/reference/mvgam-25.png index 36440c7e..89492454 100644 Binary files a/docs/reference/mvgam-25.png and b/docs/reference/mvgam-25.png differ diff --git a/docs/reference/mvgam-26.png b/docs/reference/mvgam-26.png index 2294aba4..b47b23d5 100644 Binary files a/docs/reference/mvgam-26.png and b/docs/reference/mvgam-26.png differ diff --git a/docs/reference/mvgam-27.png b/docs/reference/mvgam-27.png index 7ba58fa4..3d338ca8 100644 Binary files a/docs/reference/mvgam-27.png and b/docs/reference/mvgam-27.png differ diff --git a/docs/reference/mvgam-28.png b/docs/reference/mvgam-28.png index 31295326..626f5d23 100644 Binary files a/docs/reference/mvgam-28.png and b/docs/reference/mvgam-28.png differ diff --git a/docs/reference/mvgam-3.png b/docs/reference/mvgam-3.png index fe800df7..189c670a 100644 Binary files a/docs/reference/mvgam-3.png and b/docs/reference/mvgam-3.png differ diff --git a/docs/reference/mvgam-4.png b/docs/reference/mvgam-4.png index 3e996752..a556ce01 100644 Binary files a/docs/reference/mvgam-4.png and b/docs/reference/mvgam-4.png differ diff --git a/docs/reference/mvgam-5.png b/docs/reference/mvgam-5.png index 57ff823d..67abe00d 100644 Binary files a/docs/reference/mvgam-5.png and b/docs/reference/mvgam-5.png differ diff --git a/docs/reference/mvgam-6.png b/docs/reference/mvgam-6.png index c3ee394b..8218f902 100644 Binary files a/docs/reference/mvgam-6.png and b/docs/reference/mvgam-6.png differ diff --git a/docs/reference/mvgam-7.png b/docs/reference/mvgam-7.png index 12389767..95d23a12 100644 Binary files a/docs/reference/mvgam-7.png and b/docs/reference/mvgam-7.png differ diff --git a/docs/reference/mvgam-8.png b/docs/reference/mvgam-8.png index d50d67c1..f17f71c8 100644 Binary files a/docs/reference/mvgam-8.png and b/docs/reference/mvgam-8.png differ diff --git a/docs/reference/mvgam-9.png b/docs/reference/mvgam-9.png index 4f70efb2..195031df 100644 Binary files a/docs/reference/mvgam-9.png and b/docs/reference/mvgam-9.png differ diff --git a/docs/reference/mvgam.html b/docs/reference/mvgam.html index e902ed33..7d1df4e9 100644 --- a/docs/reference/mvgam.html +++ b/docs/reference/mvgam.html @@ -116,8 +116,8 @@

Usage algorithm = getOption("brms.algorithm", "sampling"), autoformat = TRUE, save_all_pars = FALSE, - max_treedepth = 12, - adapt_delta = 0.85, + max_treedepth = 10, + adapt_delta = 0.8, silent = 1, jags_path, ... @@ -418,7 +418,7 @@

ArgumentsExamplessdata1 <- standata(mod1) str(sdata1) #> List of 18 -#> $ y : num [1:60, 1:3] 12 17 12 31 30 8 0 1 3 5 ... +#> $ y : num [1:60, 1:3] 4 5 7 39 51 26 6 6 4 2 ... #> $ n : int 60 #> $ X : num [1:180, 1:5] 1 1 1 1 1 1 1 1 1 1 ... #> ..- attr(*, "dimnames")=List of 2 @@ -712,7 +712,7 @@

Examples#> $ zero : num [1:5] 0 0 0 0 0 #> $ p_coefs : Named num 0 #> ..- attr(*, "names")= chr "(Intercept)" -#> $ p_taus : num 1.09 +#> $ p_taus : num 0.853 #> $ ytimes : int [1:60, 1:3] 1 4 7 10 13 16 19 22 25 28 ... #> $ n_series : int 3 #> $ sp : Named num 0.368 @@ -721,10 +721,10 @@

Examples#> $ total_obs : int 180 #> $ num_basis : int 5 #> $ n_sp : num 1 -#> $ n_nonmissing: int 161 -#> $ obs_ind : int [1:161] 1 2 3 4 5 6 7 8 9 10 ... -#> $ flat_ys : num [1:161] 12 17 12 31 30 8 0 1 3 5 ... -#> $ flat_xs : num [1:161, 1:5] 1 1 1 1 1 1 1 1 1 1 ... +#> $ n_nonmissing: int 164 +#> $ obs_ind : int [1:164] 1 2 3 4 5 6 7 8 9 10 ... +#> $ flat_ys : num [1:164] 4 5 7 39 51 26 6 6 4 2 ... +#> $ flat_xs : num [1:164, 1:5] 1 1 1 1 1 1 1 1 1 1 ... #> ..- attr(*, "dimnames")=List of 2 #> .. ..$ : NULL #> .. ..$ : chr [1:5] "X.Intercept." "V2" "V3" "V4" ... @@ -746,7 +746,7 @@

Examples#> from stan/lib/stan_math/stan/math/rev.hpp:16, #> from stan/lib/stan_math/stan/math.hpp:19, #> from stan/src/stan/model/model_header.hpp:4, -#> from C:/Users/uqnclar2/AppData/Local/Temp/RtmpYHBavU/model-46c3b5e6c17.hpp:2: +#> from C:/Users/uqnclar2/AppData/Local/Temp/Rtmp0U8hqj/model-20b03ff170e8.hpp:2: #> stan/lib/stan_math/stan/math/prim/prob/von_mises_cdf.hpp: In function 'stan::return_type_t<T_x, T_sigma, T_l> stan::math::von_mises_cdf(const T_x&, const T_mu&, const T_k&)': #> stan/lib/stan_math/stan/math/prim/prob/von_mises_cdf.hpp:194: note: '-Wmisleading-indentation' is disabled from this point onwards, since column-tracking was disabled due to the size of the code/headers #> 194 | if (cdf_n < 0.0) @@ -758,18 +758,18 @@

Examples#> Chain 1 Iteration: 1 / 1000 [ 0%] (Warmup) #> Chain 2 Iteration: 1 / 1000 [ 0%] (Warmup) #> Chain 1 Iteration: 100 / 1000 [ 10%] (Warmup) -#> Chain 1 Iteration: 200 / 1000 [ 20%] (Warmup) #> Chain 2 Iteration: 100 / 1000 [ 10%] (Warmup) +#> Chain 1 Iteration: 200 / 1000 [ 20%] (Warmup) #> Chain 1 Iteration: 300 / 1000 [ 30%] (Warmup) -#> Chain 1 Iteration: 400 / 1000 [ 40%] (Warmup) #> Chain 2 Iteration: 200 / 1000 [ 20%] (Warmup) -#> Chain 2 Iteration: 300 / 1000 [ 30%] (Warmup) +#> Chain 1 Iteration: 400 / 1000 [ 40%] (Warmup) #> Chain 1 Iteration: 500 / 1000 [ 50%] (Warmup) #> Chain 1 Iteration: 501 / 1000 [ 50%] (Sampling) +#> Chain 2 Iteration: 300 / 1000 [ 30%] (Warmup) +#> Chain 1 Iteration: 600 / 1000 [ 60%] (Sampling) #> Chain 2 Iteration: 400 / 1000 [ 40%] (Warmup) #> Chain 2 Iteration: 500 / 1000 [ 50%] (Warmup) #> Chain 2 Iteration: 501 / 1000 [ 50%] (Sampling) -#> Chain 1 Iteration: 600 / 1000 [ 60%] (Sampling) #> Chain 1 Iteration: 700 / 1000 [ 70%] (Sampling) #> Chain 2 Iteration: 600 / 1000 [ 60%] (Sampling) #> Chain 1 Iteration: 800 / 1000 [ 80%] (Sampling) @@ -778,9 +778,9 @@

Examples#> Chain 2 Iteration: 800 / 1000 [ 80%] (Sampling) #> Chain 1 Iteration: 1000 / 1000 [100%] (Sampling) #> Chain 2 Iteration: 900 / 1000 [ 90%] (Sampling) -#> Chain 1 finished in 0.9 seconds. #> Chain 2 Iteration: 1000 / 1000 [100%] (Sampling) -#> Chain 2 finished in 0.9 seconds. +#> Chain 1 finished in 0.9 seconds. +#> Chain 2 finished in 1.0 seconds. #> #> Both chains finished successfully. #> Mean chain execution time: 0.9 seconds. @@ -791,7 +791,7 @@

Examplessummary(mod1) #> GAM formula: #> y ~ s(season, bs = "cc", k = 6) -#> <environment: 0x0000021ebbca2828> +#> <environment: 0x00000246410de130> #> #> Family: #> poisson @@ -816,36 +816,36 @@

Examples#> #> #> GAM coefficient (beta) estimates: -#> 2.5% 50% 97.5% Rhat n_eff -#> (Intercept) 1.80 2.000 2.1000 1.01 350 -#> s(season).1 0.15 0.370 0.6000 1.01 199 -#> s(season).2 -0.33 -0.091 0.1400 1.00 404 -#> s(season).3 -1.00 -0.710 -0.4500 1.01 372 -#> s(season).4 -0.52 -0.270 -0.0059 1.01 528 +#> 2.5% 50% 97.5% Rhat n_eff +#> (Intercept) 1.900 2.00 2.10 1 555 +#> s(season).1 0.086 0.31 0.52 1 631 +#> s(season).2 0.600 0.82 1.10 1 637 +#> s(season).3 -0.049 0.18 0.40 1 797 +#> s(season).4 -0.680 -0.42 -0.19 1 714 #> #> Approximate significance of GAM smooths: #> edf Ref.df Chi.sq p-value -#> s(season) 3.69 4 26.7 1.1e-06 *** +#> s(season) 3.52 4 39 <2e-16 *** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Latent trend parameter AR estimates: -#> 2.5% 50% 97.5% Rhat n_eff -#> ar1[1] 0.2400 0.73 0.98 1.01 274 -#> ar1[2] -0.9300 -0.36 0.18 1.01 147 -#> ar1[3] 0.0021 0.38 0.92 1.02 149 -#> sigma[1] 0.3900 0.53 0.73 1.00 314 -#> sigma[2] 0.3900 0.55 0.77 1.00 267 -#> sigma[3] 0.4300 0.59 0.78 1.00 309 +#> 2.5% 50% 97.5% Rhat n_eff +#> ar1[1] 0.32 0.73 0.990 1.01 289 +#> ar1[2] -0.93 -0.41 0.018 1.02 228 +#> ar1[3] 0.24 0.67 0.980 1.00 254 +#> sigma[1] 0.42 0.56 0.740 1.00 324 +#> sigma[2] 0.35 0.49 0.680 1.01 334 +#> sigma[3] 0.39 0.52 0.700 1.00 318 #> #> Stan MCMC diagnostics: #> n_eff / iter looks reasonable for all parameters #> Rhat looks reasonable for all parameters #> 0 of 1000 iterations ended with a divergence (0%) -#> 0 of 1000 iterations saturated the maximum tree depth of 12 (0%) +#> 0 of 1000 iterations saturated the maximum tree depth of 10 (0%) #> E-FMI indicated no pathological behavior #> -#> Samples were drawn using NUTS(diag_e) at Wed Nov 13 8:46:32 AM 2024. +#> Samples were drawn using NUTS(diag_e) at Fri Nov 15 10:09:07 AM 2024. #> For each parameter, n_eff is a crude measure of effective sample size, #> and Rhat is the potential scale reduction factor on split MCMC chains #> (at convergence, Rhat = 1) @@ -861,7 +861,7 @@

Examples resids <- residuals(mod1) str(resids) -#> num [1:180, 1:4] 0.0319 -0.06 0.1531 0.5229 -0.0881 ... +#> num [1:180, 1:4] -0.178 NaN -0.118 0.279 -0.823 ... #> - attr(*, "dimnames")=List of 2 #> ..$ : NULL #> ..$ : chr [1:4] "Estimate" "Est.Error" "Q2.5" "Q97.5" @@ -871,7 +871,7 @@

Examplesstr(fc) #> List of 16 #> $ call :Class 'formula' language y ~ s(season, bs = "cc", k = 6) -#> .. ..- attr(*, ".Environment")=<environment: 0x0000021ebbca2828> +#> .. ..- attr(*, ".Environment")=<environment: 0x00000246410de130> #> $ trend_call : NULL #> $ family : chr "poisson" #> $ family_pars : NULL @@ -890,36 +890,36 @@

Examples#> $ type : chr "response" #> $ series_names : Factor w/ 3 levels "series_1","series_2",..: 1 2 3 #> $ train_observations:List of 3 -#> ..$ series_1: int [1:60] 12 17 12 31 30 8 0 1 3 5 ... -#> ..$ series_2: int [1:60] 8 8 21 7 6 NA 1 8 9 3 ... -#> ..$ series_3: int [1:60] 10 NA 7 2 1 11 4 NA 5 4 ... +#> ..$ series_1: int [1:60] 4 5 7 39 51 26 6 6 4 2 ... +#> ..$ series_2: int [1:60] NA 2 12 16 6 31 9 15 5 3 ... +#> ..$ series_3: int [1:60] 4 NA 4 NA NA 16 7 7 3 NA ... #> $ train_times : int [1:60] 1 2 3 4 5 6 7 8 9 10 ... #> $ test_observations :List of 3 -#> ..$ series_1: int [1:20] 4 9 5 6 4 11 NA 3 8 11 ... -#> ..$ series_2: int [1:20] 19 51 19 10 3 6 7 1 6 4 ... -#> ..$ series_3: int [1:20] 10 NA 8 5 16 13 0 0 3 2 ... +#> ..$ series_1: int [1:20] 1 NA NA 13 18 20 16 6 NA 4 ... +#> ..$ series_2: int [1:20] 4 36 8 6 7 NA NA 1 6 4 ... +#> ..$ series_3: int [1:20] 6 8 5 5 19 14 1 1 7 0 ... #> $ test_times : int [1:20] 61 62 63 64 65 66 67 68 69 70 ... #> $ hindcasts :List of 3 -#> ..$ series_1: num [1:1000, 1:60] 22 9 10 1 18 15 11 19 10 15 ... +#> ..$ series_1: num [1:1000, 1:60] 4 0 1 3 0 3 4 4 3 4 ... #> .. ..- attr(*, "dimnames")=List of 2 #> .. .. ..$ : NULL #> .. .. ..$ : chr [1:60] "ypred[1,1]" "ypred[2,1]" "ypred[3,1]" "ypred[4,1]" ... -#> ..$ series_2: num [1:1000, 1:60] 12 13 5 5 12 7 16 12 6 9 ... +#> ..$ series_2: num [1:1000, 1:60] 10 5 2 0 4 20 5 8 3 7 ... #> .. ..- attr(*, "dimnames")=List of 2 #> .. .. ..$ : NULL #> .. .. ..$ : chr [1:60] "ypred[1,2]" "ypred[2,2]" "ypred[3,2]" "ypred[4,2]" ... -#> ..$ series_3: num [1:1000, 1:60] 5 11 9 9 6 15 5 6 14 12 ... +#> ..$ series_3: num [1:1000, 1:60] 6 5 4 2 3 6 6 7 6 8 ... #> .. ..- attr(*, "dimnames")=List of 2 #> .. .. ..$ : NULL #> .. .. ..$ : chr [1:60] "ypred[1,3]" "ypred[2,3]" "ypred[3,3]" "ypred[4,3]" ... #> $ forecasts :List of 3 -#> ..$ series_1: int [1:1000, 1:20] 6 8 9 2 14 4 6 4 17 6 ... -#> ..$ series_2: int [1:1000, 1:20] 10 13 12 11 8 9 1 9 25 5 ... -#> ..$ series_3: int [1:1000, 1:20] 43 18 33 11 9 26 13 10 6 3 ... +#> ..$ series_1: int [1:1000, 1:20] 4 8 0 3 10 3 1 5 2 9 ... +#> ..$ series_2: int [1:1000, 1:20] 2 5 5 0 4 1 7 5 9 13 ... +#> ..$ series_3: int [1:1000, 1:20] 7 3 6 10 3 6 6 3 4 11 ... #> - attr(*, "class")= chr "mvgam_forecast" plot(fc) #> Out of sample DRPS: -#> 55.270817 +#> 57.561702 # Plot the estimated seasonal smooth function @@ -943,7 +943,7 @@

Examplesconditional_effects(mod1) plot_predictions(mod1, condition = 'season', points = 0.5) -#> Warning: Removed 19 rows containing missing values or values outside the scale range +#> Warning: Removed 16 rows containing missing values or values outside the scale range #> (`geom_point()`). @@ -957,19 +957,19 @@

Examplesbeta_draws_df <- as.data.frame(mod1, variable = 'betas') head(beta_draws_df) #> (Intercept) s(season).1 s(season).2 s(season).3 s(season).4 -#> 1 1.94416 0.394466 -0.06900860 -0.783359 -0.426288 -#> 2 1.92798 0.271471 -0.00850006 -0.685065 -0.244868 -#> 3 1.90193 0.230704 -0.04661740 -0.676646 -0.314741 -#> 4 1.88637 0.430494 -0.30333200 -0.634353 -0.345208 -#> 5 1.97914 0.322658 0.15346600 -0.703236 -0.263105 -#> 6 2.02798 0.407478 0.05122050 -0.581718 -0.361806 +#> 1 1.93652 0.335460 0.781389 -0.0251807 -0.423511 +#> 2 2.02126 0.389714 0.795744 0.1534600 -0.584915 +#> 3 1.98538 0.232350 0.944969 0.1425350 -0.444939 +#> 4 1.89741 0.300367 0.785340 0.2458830 -0.414404 +#> 5 1.99933 0.352974 0.763884 0.1705710 -0.261901 +#> 6 1.99175 0.229309 0.742755 0.1530400 -0.495284 str(beta_draws_df) #> 'data.frame': 1000 obs. of 5 variables: -#> $ (Intercept): num 1.94 1.93 1.9 1.89 1.98 ... -#> $ s(season).1: num 0.394 0.271 0.231 0.43 0.323 ... -#> $ s(season).2: num -0.069 -0.0085 -0.0466 -0.3033 0.1535 ... -#> $ s(season).3: num -0.783 -0.685 -0.677 -0.634 -0.703 ... -#> $ s(season).4: num -0.426 -0.245 -0.315 -0.345 -0.263 ... +#> $ (Intercept): num 1.94 2.02 1.99 1.9 2 ... +#> $ s(season).1: num 0.335 0.39 0.232 0.3 0.353 ... +#> $ s(season).2: num 0.781 0.796 0.945 0.785 0.764 ... +#> $ s(season).3: num -0.0252 0.1535 0.1425 0.2459 0.1706 ... +#> $ s(season).4: num -0.424 -0.585 -0.445 -0.414 -0.262 ... # Investigate model fit mc.cores.def <- getOption('mc.cores') @@ -977,21 +977,21 @@

Examplesloo(mod1) #> Warning: Some Pareto k diagnostic values are too high. See help('pareto-k-diagnostic') for details. #> -#> Computed from 1000 by 161 log-likelihood matrix. +#> Computed from 1000 by 164 log-likelihood matrix. #> #> Estimate SE -#> elpd_loo -452.2 8.5 -#> p_loo 91.6 4.8 -#> looic 904.4 17.0 +#> elpd_loo -458.4 9.2 +#> p_loo 88.8 5.2 +#> looic 916.9 18.5 #> ------ #> MCSE of elpd_loo is NA. -#> MCSE and ESS estimates assume MCMC draws (r_eff in [0.3, 1.6]). +#> MCSE and ESS estimates assume MCMC draws (r_eff in [0.3, 1.8]). #> #> Pareto k diagnostic values: #> Count Pct. Min. ESS -#> (-Inf, 0.67] (good) 67 41.6% 85 -#> (0.67, 1] (bad) 79 49.1% <NA> -#> (1, Inf) (very bad) 15 9.3% <NA> +#> (-Inf, 0.67] (good) 77 47.0% 78 +#> (0.67, 1] (bad) 77 47.0% <NA> +#> (1, Inf) (very bad) 10 6.1% <NA> #> See help('pareto-k-diagnostic') for details. options(mc.cores = mc.cores.def) @@ -1022,7 +1022,7 @@

Examples#> from stan/lib/stan_math/stan/math/rev.hpp:16, #> from stan/lib/stan_math/stan/math.hpp:19, #> from stan/src/stan/model/model_header.hpp:4, -#> from C:/Users/uqnclar2/AppData/Local/Temp/RtmpYHBavU/model-46c47d355d4.hpp:2: +#> from C:/Users/uqnclar2/AppData/Local/Temp/Rtmp0U8hqj/model-20b0192b6bd2.hpp:2: #> stan/lib/stan_math/stan/math/prim/prob/von_mises_cdf.hpp: In function 'stan::return_type_t<T_x, T_sigma, T_l> stan::math::von_mises_cdf(const T_x&, const T_mu&, const T_k&)': #> stan/lib/stan_math/stan/math/prim/prob/von_mises_cdf.hpp:194: note: '-Wmisleading-indentation' is disabled from this point onwards, since column-tracking was disabled due to the size of the code/headers #> 194 | if (cdf_n < 0.0) @@ -1035,32 +1035,32 @@

Examples#> Chain 1 Iteration: 100 / 1000 [ 10%] (Warmup) #> Chain 2 Iteration: 1 / 1000 [ 0%] (Warmup) #> Chain 1 Iteration: 200 / 1000 [ 20%] (Warmup) -#> Chain 1 Iteration: 300 / 1000 [ 30%] (Warmup) -#> Chain 1 Iteration: 400 / 1000 [ 40%] (Warmup) #> Chain 2 Iteration: 100 / 1000 [ 10%] (Warmup) #> Chain 2 Iteration: 200 / 1000 [ 20%] (Warmup) -#> Chain 1 Iteration: 500 / 1000 [ 50%] (Warmup) -#> Chain 1 Iteration: 501 / 1000 [ 50%] (Sampling) +#> Chain 1 Iteration: 300 / 1000 [ 30%] (Warmup) #> Chain 2 Iteration: 300 / 1000 [ 30%] (Warmup) -#> Chain 1 Iteration: 600 / 1000 [ 60%] (Sampling) +#> Chain 1 Iteration: 400 / 1000 [ 40%] (Warmup) #> Chain 2 Iteration: 400 / 1000 [ 40%] (Warmup) -#> Chain 1 Iteration: 700 / 1000 [ 70%] (Sampling) -#> Chain 1 Iteration: 800 / 1000 [ 80%] (Sampling) +#> Chain 1 Iteration: 500 / 1000 [ 50%] (Warmup) +#> Chain 1 Iteration: 501 / 1000 [ 50%] (Sampling) #> Chain 2 Iteration: 500 / 1000 [ 50%] (Warmup) #> Chain 2 Iteration: 501 / 1000 [ 50%] (Sampling) -#> Chain 1 Iteration: 900 / 1000 [ 90%] (Sampling) +#> Chain 1 Iteration: 600 / 1000 [ 60%] (Sampling) +#> Chain 1 Iteration: 700 / 1000 [ 70%] (Sampling) #> Chain 2 Iteration: 600 / 1000 [ 60%] (Sampling) -#> Chain 1 Iteration: 1000 / 1000 [100%] (Sampling) -#> Chain 1 finished in 1.1 seconds. #> Chain 2 Iteration: 700 / 1000 [ 70%] (Sampling) +#> Chain 1 Iteration: 800 / 1000 [ 80%] (Sampling) #> Chain 2 Iteration: 800 / 1000 [ 80%] (Sampling) +#> Chain 1 Iteration: 900 / 1000 [ 90%] (Sampling) #> Chain 2 Iteration: 900 / 1000 [ 90%] (Sampling) +#> Chain 1 Iteration: 1000 / 1000 [100%] (Sampling) +#> Chain 1 finished in 1.4 seconds. #> Chain 2 Iteration: 1000 / 1000 [100%] (Sampling) -#> Chain 2 finished in 2.0 seconds. +#> Chain 2 finished in 1.4 seconds. #> #> Both chains finished successfully. -#> Mean chain execution time: 1.6 seconds. -#> Total execution time: 2.2 seconds. +#> Mean chain execution time: 1.4 seconds. +#> Total execution time: 1.6 seconds. #> # The mapping matrix is now supplied as data to the model in the 'Z' element @@ -1124,7 +1124,7 @@

Examples#> } #> model { #> // prior for (Intercept)... -#> b_raw[1] ~ student_t(3, 0, 2.5); +#> b_raw[1] ~ student_t(3, -2.3, 2.5); #> #> // prior for s(season)... #> b_raw[2 : 5] ~ multi_normal_prec(zero[2 : 5], S1[1 : 4, 1 : 4] * lambda[1]); @@ -1213,6 +1213,9 @@

Examples data = data_train, newdata = data_test, chains = 2) +#> Warning: gp effects in mvgam cannot yet handle autogrouping +#> resetting all instances of 'gr = TRUE' to 'gr = FALSE' +#> This warning is displayed once per session. #> Compiling Stan program using cmdstanr #> #> In file included from stan/lib/stan_math/stan/math/prim/prob/von_mises_lccdf.hpp:5, @@ -1222,7 +1225,7 @@

Examples#> from stan/lib/stan_math/stan/math/rev.hpp:16, #> from stan/lib/stan_math/stan/math.hpp:19, #> from stan/src/stan/model/model_header.hpp:4, -#> from C:/Users/uqnclar2/AppData/Local/Temp/RtmpYHBavU/model-46c9b418f5.hpp:2: +#> from C:/Users/uqnclar2/AppData/Local/Temp/Rtmp0U8hqj/model-20b0744353df.hpp:2: #> stan/lib/stan_math/stan/math/prim/prob/von_mises_cdf.hpp: In function 'stan::return_type_t<T_x, T_sigma, T_l> stan::math::von_mises_cdf(const T_x&, const T_mu&, const T_k&)': #> stan/lib/stan_math/stan/math/prim/prob/von_mises_cdf.hpp:194: note: '-Wmisleading-indentation' is disabled from this point onwards, since column-tracking was disabled due to the size of the code/headers #> 194 | if (cdf_n < 0.0) @@ -1255,11 +1258,11 @@

Examples#> Chain 2 Iteration: 900 / 1000 [ 90%] (Sampling) #> Chain 1 Iteration: 1000 / 1000 [100%] (Sampling) #> Chain 2 Iteration: 1000 / 1000 [100%] (Sampling) -#> Chain 1 finished in 0.8 seconds. -#> Chain 2 finished in 0.8 seconds. +#> Chain 1 finished in 0.7 seconds. +#> Chain 2 finished in 0.7 seconds. #> #> Both chains finished successfully. -#> Mean chain execution time: 0.8 seconds. +#> Mean chain execution time: 0.7 seconds. #> Total execution time: 0.9 seconds. #> @@ -1267,7 +1270,7 @@

Examplessummary(mod) #> GAM formula: #> out ~ gp(time, by = temp, c = 5/4, k = 40, scale = FALSE) -#> <environment: 0x0000021ebbca2828> +#> <environment: 0x00000246410de130> #> #> Family: #> gaussian @@ -1292,66 +1295,66 @@

Examples#> #> Observation error parameter estimates: #> 2.5% 50% 97.5% Rhat n_eff -#> sigma_obs[1] 0.44 0.49 0.55 1 1301 +#> sigma_obs[1] 0.44 0.49 0.55 1 1810 #> #> GAM coefficient (beta) estimates: -#> 2.5% 50% 97.5% Rhat n_eff -#> (Intercept) 4.000 4.0e+00 4.100 1.00 1985 -#> gp(time):temp.1 0.430 3.2e+00 6.700 1.00 962 -#> gp(time):temp.2 -3.300 1.5e+00 6.500 1.00 969 -#> gp(time):temp.3 -6.200 -1.4e+00 3.300 1.00 1060 -#> gp(time):temp.4 -5.700 -1.5e+00 2.800 1.00 952 -#> gp(time):temp.5 -3.700 2.0e-01 3.800 1.00 1123 -#> gp(time):temp.6 -2.500 6.1e-01 4.700 1.00 1104 -#> gp(time):temp.7 -3.200 -1.9e-01 2.900 1.00 831 -#> gp(time):temp.8 -2.700 -4.8e-02 2.300 1.00 1025 -#> gp(time):temp.9 -1.600 5.8e-01 3.200 1.00 863 -#> gp(time):temp.10 -2.100 -5.0e-02 1.800 1.00 1422 -#> gp(time):temp.11 -2.500 -4.5e-01 1.000 1.00 904 -#> gp(time):temp.12 -1.400 7.1e-02 1.900 1.00 1162 -#> gp(time):temp.13 -0.890 1.7e-01 1.900 1.00 1547 -#> gp(time):temp.14 -1.600 -2.3e-02 1.200 1.00 1174 -#> gp(time):temp.15 -1.600 -1.6e-02 1.000 1.00 1062 -#> gp(time):temp.16 -0.850 1.7e-03 1.500 1.00 1316 -#> gp(time):temp.17 -0.950 4.3e-04 1.300 1.00 1338 -#> gp(time):temp.18 -1.100 -1.8e-15 0.910 1.00 1409 -#> gp(time):temp.19 -0.980 -2.4e-08 0.930 1.00 1809 -#> gp(time):temp.20 -0.970 -3.8e-05 0.690 1.00 1603 -#> gp(time):temp.21 -0.510 3.2e-06 0.860 1.00 1660 -#> gp(time):temp.22 -0.370 3.0e-04 0.950 1.00 799 -#> gp(time):temp.23 -0.700 -5.4e-08 0.480 1.00 1358 -#> gp(time):temp.24 -0.850 -6.9e-07 0.290 1.00 939 -#> gp(time):temp.25 -0.330 5.3e-10 0.680 1.01 872 -#> gp(time):temp.26 -0.380 -6.0e-40 0.560 1.00 1064 -#> gp(time):temp.27 -0.600 -1.6e-09 0.260 1.00 1177 -#> gp(time):temp.28 -0.460 -1.6e-23 0.460 1.00 1112 -#> gp(time):temp.29 -0.230 2.7e-12 0.490 1.00 895 -#> gp(time):temp.30 -0.280 -2.5e-40 0.230 1.00 1522 -#> gp(time):temp.31 -0.610 -2.6e-21 0.200 1.00 934 -#> gp(time):temp.32 -0.240 3.5e-33 0.350 1.00 1125 -#> gp(time):temp.33 -0.200 5.5e-56 0.330 1.00 1734 -#> gp(time):temp.34 -0.210 -4.6e-36 0.150 1.00 1047 -#> gp(time):temp.35 -0.210 1.2e-44 0.230 1.00 1183 -#> gp(time):temp.36 -0.180 2.6e-29 0.190 1.00 1430 -#> gp(time):temp.37 -0.120 -4.7e-53 0.180 1.00 1225 -#> gp(time):temp.38 -0.160 -3.0e-57 0.082 1.00 1136 -#> gp(time):temp.39 -0.180 -5.8e-37 0.068 1.00 964 -#> gp(time):temp.40 -0.043 9.7e-43 0.160 1.00 728 +#> 2.5% 50% 97.5% Rhat n_eff +#> (Intercept) 4.000 4.0e+00 4.100 1.00 1293 +#> gp(time):temp.1 0.089 3.1e+00 7.100 1.00 710 +#> gp(time):temp.2 -3.700 1.6e+00 7.100 1.00 535 +#> gp(time):temp.3 -6.900 -1.5e+00 3.500 1.00 879 +#> gp(time):temp.4 -6.100 -1.5e+00 2.700 1.00 477 +#> gp(time):temp.5 -3.800 2.1e-01 4.300 1.00 899 +#> gp(time):temp.6 -3.000 4.9e-01 4.100 1.00 710 +#> gp(time):temp.7 -3.700 -1.9e-01 3.400 1.00 353 +#> gp(time):temp.8 -3.100 -1.8e-02 2.300 1.00 1068 +#> gp(time):temp.9 -1.700 4.8e-01 3.400 1.00 432 +#> gp(time):temp.10 -2.400 -4.7e-03 2.400 1.00 699 +#> gp(time):temp.11 -2.500 -3.1e-01 0.820 1.00 712 +#> gp(time):temp.12 -1.200 4.0e-02 2.000 1.00 882 +#> gp(time):temp.13 -0.890 6.6e-02 2.000 1.00 777 +#> gp(time):temp.14 -1.500 -7.0e-03 1.100 1.00 1288 +#> gp(time):temp.15 -1.400 -5.0e-04 0.920 1.00 1238 +#> gp(time):temp.16 -0.710 1.3e-05 1.300 1.00 1092 +#> gp(time):temp.17 -0.710 2.3e-05 0.980 1.00 1305 +#> gp(time):temp.18 -0.940 -1.9e-11 0.790 1.00 1080 +#> gp(time):temp.19 -0.760 -1.4e-10 0.670 1.00 1266 +#> gp(time):temp.20 -0.910 -3.7e-09 0.460 1.00 976 +#> gp(time):temp.21 -0.490 -1.0e-14 0.660 1.00 1346 +#> gp(time):temp.22 -0.260 9.3e-08 0.820 1.00 1068 +#> gp(time):temp.23 -0.600 -7.3e-20 0.390 1.00 1091 +#> gp(time):temp.24 -0.540 -1.1e-08 0.190 1.01 722 +#> gp(time):temp.25 -0.200 8.3e-19 0.480 1.00 1042 +#> gp(time):temp.26 -0.240 1.2e-42 0.400 1.00 1194 +#> gp(time):temp.27 -0.430 -1.5e-18 0.130 1.00 900 +#> gp(time):temp.28 -0.230 -3.0e-56 0.250 1.00 1346 +#> gp(time):temp.29 -0.110 5.1e-24 0.430 1.00 973 +#> gp(time):temp.30 -0.250 -6.4e-35 0.160 1.00 1348 +#> gp(time):temp.31 -0.310 -1.8e-40 0.091 1.00 1028 +#> gp(time):temp.32 -0.140 -9.4e-74 0.130 1.01 827 +#> gp(time):temp.33 -0.063 3.0e-87 0.200 1.00 1123 +#> gp(time):temp.34 -0.160 -9.0e-49 0.068 1.00 1029 +#> gp(time):temp.35 -0.060 2.3e-42 0.120 1.00 1154 +#> gp(time):temp.36 -0.080 -9.4e-64 0.076 1.00 1226 +#> gp(time):temp.37 -0.076 -1.9e-41 0.075 1.00 1410 +#> gp(time):temp.38 -0.110 -4.3e-43 0.034 1.00 1029 +#> gp(time):temp.39 -0.041 2.1e-116 0.052 1.00 1174 +#> gp(time):temp.40 -0.016 3.7e-52 0.058 1.00 871 #> #> GAM gp term marginal deviation (alpha) and length scale (rho) estimates: -#> 2.5% 50% 97.5% Rhat n_eff -#> alpha_gp(time):temp 0.18 0.33 0.78 1.00 508 -#> rho_gp(time):temp 9.30 29.00 90.00 1.01 382 +#> 2.5% 50% 97.5% Rhat n_eff +#> alpha_gp(time):temp 0.17 0.34 0.83 1 273 +#> rho_gp(time):temp 12.00 32.00 100.00 1 305 #> #> Stan MCMC diagnostics: #> n_eff / iter looks reasonable for all parameters #> Rhat looks reasonable for all parameters -#> 29 of 1000 iterations ended with a divergence (2.9%) +#> 33 of 1000 iterations ended with a divergence (3.3%) #> *Try running with larger adapt_delta to remove the divergences -#> 0 of 1000 iterations saturated the maximum tree depth of 12 (0%) +#> 0 of 1000 iterations saturated the maximum tree depth of 10 (0%) #> E-FMI indicated no pathological behavior #> -#> Samples were drawn using NUTS(diag_e) at Wed Nov 13 8:47:47 AM 2024. +#> Samples were drawn using NUTS(diag_e) at Fri Nov 15 10:10:13 AM 2024. #> For each parameter, n_eff is a crude measure of effective sample size, #> and Rhat is the potential scale reduction factor on split MCMC chains #> (at convergence, Rhat = 1) @@ -1360,7 +1363,7 @@

Examplesfc <- forecast(mod, newdata = data_test) plot(fc) #> Out of sample CRPS: -#> 6.43188275110833 +#> 6.31126950637918 # Propagating the smooth term shows how the coefficient is expected to evolve @@ -1397,7 +1400,7 @@

Examples#> from stan/lib/stan_math/stan/math/rev.hpp:16, #> from stan/lib/stan_math/stan/math.hpp:19, #> from stan/src/stan/model/model_header.hpp:4, -#> from C:/Users/uqnclar2/AppData/Local/Temp/RtmpYHBavU/model-46c3dc259d2.hpp:2: +#> from C:/Users/uqnclar2/AppData/Local/Temp/Rtmp0U8hqj/model-20b06c442a55.hpp:2: #> stan/lib/stan_math/stan/math/prim/prob/von_mises_cdf.hpp: In function 'stan::return_type_t<T_x, T_sigma, T_l> stan::math::von_mises_cdf(const T_x&, const T_mu&, const T_k&)': #> stan/lib/stan_math/stan/math/prim/prob/von_mises_cdf.hpp:194: note: '-Wmisleading-indentation' is disabled from this point onwards, since column-tracking was disabled due to the size of the code/headers #> 194 | if (cdf_n < 0.0) @@ -1410,32 +1413,32 @@

Examples#> Chain 2 Iteration: 1 / 1000 [ 0%] (Warmup) #> Chain 1 Iteration: 100 / 1000 [ 10%] (Warmup) #> Chain 2 Iteration: 100 / 1000 [ 10%] (Warmup) -#> Chain 2 Iteration: 200 / 1000 [ 20%] (Warmup) #> Chain 1 Iteration: 200 / 1000 [ 20%] (Warmup) -#> Chain 2 Iteration: 300 / 1000 [ 30%] (Warmup) -#> Chain 2 Iteration: 400 / 1000 [ 40%] (Warmup) +#> Chain 2 Iteration: 200 / 1000 [ 20%] (Warmup) #> Chain 1 Iteration: 300 / 1000 [ 30%] (Warmup) -#> Chain 2 Iteration: 500 / 1000 [ 50%] (Warmup) -#> Chain 2 Iteration: 501 / 1000 [ 50%] (Sampling) +#> Chain 2 Iteration: 300 / 1000 [ 30%] (Warmup) #> Chain 1 Iteration: 400 / 1000 [ 40%] (Warmup) -#> Chain 2 Iteration: 600 / 1000 [ 60%] (Sampling) +#> Chain 2 Iteration: 400 / 1000 [ 40%] (Warmup) #> Chain 1 Iteration: 500 / 1000 [ 50%] (Warmup) #> Chain 1 Iteration: 501 / 1000 [ 50%] (Sampling) -#> Chain 2 Iteration: 700 / 1000 [ 70%] (Sampling) +#> Chain 2 Iteration: 500 / 1000 [ 50%] (Warmup) +#> Chain 2 Iteration: 501 / 1000 [ 50%] (Sampling) #> Chain 1 Iteration: 600 / 1000 [ 60%] (Sampling) -#> Chain 2 Iteration: 800 / 1000 [ 80%] (Sampling) +#> Chain 2 Iteration: 600 / 1000 [ 60%] (Sampling) #> Chain 1 Iteration: 700 / 1000 [ 70%] (Sampling) -#> Chain 2 Iteration: 900 / 1000 [ 90%] (Sampling) -#> Chain 2 Iteration: 1000 / 1000 [100%] (Sampling) -#> Chain 2 finished in 3.8 seconds. +#> Chain 2 Iteration: 700 / 1000 [ 70%] (Sampling) #> Chain 1 Iteration: 800 / 1000 [ 80%] (Sampling) #> Chain 1 Iteration: 900 / 1000 [ 90%] (Sampling) +#> Chain 2 Iteration: 800 / 1000 [ 80%] (Sampling) #> Chain 1 Iteration: 1000 / 1000 [100%] (Sampling) -#> Chain 1 finished in 5.0 seconds. +#> Chain 1 finished in 3.6 seconds. +#> Chain 2 Iteration: 900 / 1000 [ 90%] (Sampling) +#> Chain 2 Iteration: 1000 / 1000 [100%] (Sampling) +#> Chain 2 finished in 4.0 seconds. #> #> Both chains finished successfully. -#> Mean chain execution time: 4.4 seconds. -#> Total execution time: 5.1 seconds. +#> Mean chain execution time: 3.8 seconds. +#> Total execution time: 4.2 seconds. #> # Inspect the model file to see the modification to the linear predictor @@ -1488,7 +1491,7 @@

Examples#> mu_raw ~ std_normal(); #> #> // prior for (Intercept)... -#> b_raw[1] ~ student_t(3, 1.4, 2.5); +#> b_raw[1] ~ student_t(3, 1.6, 2.5); #> #> // prior for s(season)... #> b_raw[2 : 9] ~ multi_normal_prec(zero[2 : 9], S1[1 : 8, 1 : 8] * lambda[1]); @@ -1537,10 +1540,10 @@

Exampleslayout(matrix(1:2, ncol = 2)) plot(fc, series = 1, ylim = c(0, 75)) #> Out of sample DRPS: -#> 22.781273 +#> 26.747011 plot(fc, series = 2, ylim = c(0, 75)) #> Out of sample DRPS: -#> 77.61833 +#> 101.610568 layout(1) @@ -1550,7 +1553,7 @@

Examplesfc <- forecast(mod, newdata = dat$data_test) plot(fc) #> Out of sample DRPS: -#> 38.523539 +#> 41.417288 # Relative Risks can be computed by fixing the offset to the same value @@ -1619,7 +1622,7 @@

Examples#> from stan/lib/stan_math/stan/math/rev.hpp:16, #> from stan/lib/stan_math/stan/math.hpp:19, #> from stan/src/stan/model/model_header.hpp:4, -#> from C:/Users/uqnclar2/AppData/Local/Temp/RtmpYHBavU/model-46c5b973bb6.hpp:2: +#> from C:/Users/uqnclar2/AppData/Local/Temp/Rtmp0U8hqj/model-20b02cf01a2b.hpp:2: #> stan/lib/stan_math/stan/math/prim/prob/von_mises_cdf.hpp: In function 'stan::return_type_t<T_x, T_sigma, T_l> stan::math::von_mises_cdf(const T_x&, const T_mu&, const T_k&)': #> stan/lib/stan_math/stan/math/prim/prob/von_mises_cdf.hpp:194: note: '-Wmisleading-indentation' is disabled from this point onwards, since column-tracking was disabled due to the size of the code/headers #> 194 | if (cdf_n < 0.0) @@ -1633,27 +1636,27 @@

Examples#> Chain 1 Iteration: 100 / 1000 [ 10%] (Warmup) #> Chain 2 Iteration: 100 / 1000 [ 10%] (Warmup) #> Chain 1 Iteration: 200 / 1000 [ 20%] (Warmup) -#> Chain 2 Iteration: 200 / 1000 [ 20%] (Warmup) #> Chain 1 Iteration: 300 / 1000 [ 30%] (Warmup) +#> Chain 2 Iteration: 200 / 1000 [ 20%] (Warmup) #> Chain 2 Iteration: 300 / 1000 [ 30%] (Warmup) -#> Chain 2 Iteration: 400 / 1000 [ 40%] (Warmup) #> Chain 1 Iteration: 400 / 1000 [ 40%] (Warmup) #> Chain 1 Iteration: 500 / 1000 [ 50%] (Warmup) #> Chain 1 Iteration: 501 / 1000 [ 50%] (Sampling) +#> Chain 2 Iteration: 400 / 1000 [ 40%] (Warmup) +#> Chain 1 Iteration: 600 / 1000 [ 60%] (Sampling) #> Chain 2 Iteration: 500 / 1000 [ 50%] (Warmup) #> Chain 2 Iteration: 501 / 1000 [ 50%] (Sampling) -#> Chain 1 Iteration: 600 / 1000 [ 60%] (Sampling) #> Chain 2 Iteration: 600 / 1000 [ 60%] (Sampling) -#> Chain 2 Iteration: 700 / 1000 [ 70%] (Sampling) #> Chain 1 Iteration: 700 / 1000 [ 70%] (Sampling) -#> Chain 2 Iteration: 800 / 1000 [ 80%] (Sampling) #> Chain 1 Iteration: 800 / 1000 [ 80%] (Sampling) -#> Chain 2 Iteration: 900 / 1000 [ 90%] (Sampling) +#> Chain 2 Iteration: 700 / 1000 [ 70%] (Sampling) #> Chain 1 Iteration: 900 / 1000 [ 90%] (Sampling) -#> Chain 2 Iteration: 1000 / 1000 [100%] (Sampling) -#> Chain 2 finished in 0.9 seconds. +#> Chain 2 Iteration: 800 / 1000 [ 80%] (Sampling) #> Chain 1 Iteration: 1000 / 1000 [100%] (Sampling) -#> Chain 1 finished in 1.1 seconds. +#> Chain 2 Iteration: 900 / 1000 [ 90%] (Sampling) +#> Chain 1 finished in 0.9 seconds. +#> Chain 2 Iteration: 1000 / 1000 [100%] (Sampling) +#> Chain 2 finished in 1.0 seconds. #> #> Both chains finished successfully. #> Mean chain execution time: 1.0 seconds. @@ -1662,7 +1665,7 @@

Examplessummary(mod) #> GAM formula: #> cbind(y, ntrials) ~ series + s(x, by = series) -#> <environment: 0x0000021ebbca2828> +#> <environment: 0x00000246410de130> #> #> Family: #> binomial @@ -1687,42 +1690,44 @@

Examples#> #> GAM coefficient (beta) estimates: #> 2.5% 50% 97.5% Rhat n_eff -#> (Intercept) -0.280 -0.17000 -0.036 1.00 592 -#> seriesseries2 -0.320 -0.15000 0.028 1.00 577 -#> s(x):seriesseries1.1 -0.120 0.02100 0.380 1.01 176 -#> s(x):seriesseries1.2 -0.190 0.00420 0.260 1.00 222 -#> s(x):seriesseries1.3 -0.088 -0.00440 0.063 1.00 307 -#> s(x):seriesseries1.4 -0.150 -0.00084 0.150 1.00 250 -#> s(x):seriesseries1.5 -0.064 0.00330 0.092 1.00 232 -#> s(x):seriesseries1.6 -0.120 -0.00120 0.130 1.00 253 -#> s(x):seriesseries1.7 -0.036 -0.00051 0.028 1.00 364 -#> s(x):seriesseries1.8 -0.440 0.00740 0.500 1.00 223 -#> s(x):seriesseries1.9 0.280 0.70000 0.920 1.01 171 -#> s(x):seriesseries2.1 -0.650 -0.07300 0.070 1.02 76 -#> s(x):seriesseries2.2 -0.270 0.02900 0.520 1.03 138 -#> s(x):seriesseries2.3 -0.068 0.01300 0.170 1.02 191 -#> s(x):seriesseries2.4 -0.160 0.02600 0.340 1.03 142 -#> s(x):seriesseries2.5 -0.180 -0.01700 0.055 1.03 127 -#> s(x):seriesseries2.6 -0.120 0.02400 0.310 1.03 130 -#> s(x):seriesseries2.7 -0.031 0.00420 0.079 1.00 374 -#> s(x):seriesseries2.8 -0.430 0.12000 1.200 1.03 117 -#> s(x):seriesseries2.9 -0.800 -0.58000 0.013 1.02 89 +#> (Intercept) -0.540 -0.41000 -0.260 1.00 679 +#> seriesseries2 0.007 0.19000 0.370 1.00 725 +#> s(x):seriesseries1.1 -0.160 0.04300 0.800 1.06 36 +#> s(x):seriesseries1.2 -0.540 -0.01500 0.480 1.00 136 +#> s(x):seriesseries1.3 -0.430 -0.00990 0.086 1.05 42 +#> s(x):seriesseries1.4 -0.220 -0.00910 0.290 1.00 122 +#> s(x):seriesseries1.5 -0.150 0.00330 0.100 1.00 126 +#> s(x):seriesseries1.6 -0.270 0.01100 0.210 1.00 123 +#> s(x):seriesseries1.7 -0.098 -0.00370 0.110 1.00 128 +#> s(x):seriesseries1.8 -1.300 -0.08100 1.100 1.00 144 +#> s(x):seriesseries1.9 0.034 1.10000 1.400 1.06 36 +#> s(x):seriesseries2.1 -0.320 -0.03100 0.140 1.00 209 +#> s(x):seriesseries2.2 -0.230 0.01100 0.270 1.00 201 +#> s(x):seriesseries2.3 -0.083 0.00850 0.130 1.00 379 +#> s(x):seriesseries2.4 -0.110 0.00150 0.140 1.00 234 +#> s(x):seriesseries2.5 -0.062 0.00180 0.063 1.00 339 +#> s(x):seriesseries2.6 -0.120 -0.00034 0.110 1.00 262 +#> s(x):seriesseries2.7 -0.045 -0.00015 0.053 1.00 273 +#> s(x):seriesseries2.8 -0.550 0.00500 0.690 1.00 228 +#> s(x):seriesseries2.9 -1.000 -0.77000 -0.390 1.00 197 #> #> Approximate significance of GAM smooths: #> edf Ref.df Chi.sq p-value -#> s(x):seriesseries1 2.12 9 27.4 < 2e-16 *** -#> s(x):seriesseries2 1.77 9 38.8 0.00017 *** +#> s(x):seriesseries1 2.04 9 88.8 <2e-16 *** +#> s(x):seriesseries2 1.17 9 34.3 2e-04 *** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Stan MCMC diagnostics: #> n_eff / iter looks reasonable for all parameters -#> Rhat looks reasonable for all parameters -#> 0 of 1000 iterations ended with a divergence (0%) -#> 0 of 1000 iterations saturated the maximum tree depth of 12 (0%) +#> Rhats above 1.05 found for 3 parameters +#> *Diagnose further to investigate why the chains have not mixed +#> 1 of 1000 iterations ended with a divergence (0.1%) +#> *Try running with larger adapt_delta to remove the divergences +#> 0 of 1000 iterations saturated the maximum tree depth of 10 (0%) #> E-FMI indicated no pathological behavior #> -#> Samples were drawn using NUTS(diag_e) at Wed Nov 13 8:48:51 AM 2024. +#> Samples were drawn using NUTS(diag_e) at Fri Nov 15 10:11:16 AM 2024. #> For each parameter, n_eff is a crude measure of effective sample size, #> and Rhat is the potential scale reduction factor on split MCMC chains #> (at convergence, Rhat = 1) diff --git a/docs/reference/sim_mvgam-1.png b/docs/reference/sim_mvgam-1.png index 8df34a89..a1097f1d 100644 Binary files a/docs/reference/sim_mvgam-1.png and b/docs/reference/sim_mvgam-1.png differ diff --git a/docs/reference/sim_mvgam-2.png b/docs/reference/sim_mvgam-2.png index ceda0f49..ccb321da 100644 Binary files a/docs/reference/sim_mvgam-2.png and b/docs/reference/sim_mvgam-2.png differ diff --git a/docs/reference/sim_mvgam.html b/docs/reference/sim_mvgam.html index f641d637..f5e76735 100644 --- a/docs/reference/sim_mvgam.html +++ b/docs/reference/sim_mvgam.html @@ -1,9 +1,11 @@ Simulate a set of time series for mvgam modelling — sim_mvgam • mvgam @@ -63,8 +65,9 @@

This function simulates sets of time series data for fitting a multivariate GAM that includes -shared seasonality and dependence on state-space latent dynamic factors. Random dependencies among series, i.e. -correlations in their long-term trends, are included in the form of correlated loadings on the latent dynamic factors

+shared seasonality and dependence on state-space latent dynamic factors. Random +dependencies among series, i.e. correlations in their long-term trends, are included +in the form of correlated loadings on the latent dynamic factors

@@ -102,13 +105,16 @@

ArgumentsArgumentsgam)

+

character specifying the time series dynamics for the latent trend. +Options are:

@@ -184,25 +194,30 @@

Arguments

Value

-

A list object containing outputs needed for mvgam, including 'data_train' and 'data_test', -as well as some additional information about the simulated seasonality and trend dependencies

+

A list object containing outputs needed for mvgam, +including 'data_train' and 'data_test', as well as some additional information +about the simulated seasonality and trend dependencies

diff --git a/man/forecast.mvgam.Rd b/man/forecast.mvgam.Rd index 45070d31..0291f212 100644 --- a/man/forecast.mvgam.Rd +++ b/man/forecast.mvgam.Rd @@ -61,7 +61,8 @@ mod <- mvgam(y ~ s(season, bs = 'cc', k = 6), trend_model = AR(), noncentred = TRUE, data = simdat$data_train, - chains = 2) + chains = 2, + silent = 2) # Hindcasts on response scale hc <- hindcast(mod) diff --git a/man/lfo_cv.mvgam.Rd b/man/lfo_cv.mvgam.Rd index ea6b0de3..10d61aa9 100644 --- a/man/lfo_cv.mvgam.Rd +++ b/man/lfo_cv.mvgam.Rd @@ -98,7 +98,8 @@ mod_ar2 <- mvgam(y ~ s(season, bs = 'cc', k = 6), newdata = dat$data_test, burnin = 300, samples = 300, - chains = 2) + chains = 2, + silent = 2) # Fit a less appropriate model mod_rw <- mvgam(y ~ s(season, bs = 'cc', k = 6), @@ -108,7 +109,8 @@ mod_rw <- mvgam(y ~ s(season, bs = 'cc', k = 6), newdata = dat$data_test, burnin = 300, samples = 300, - chains = 2) + chains = 2, + silent = 2) # Compare Discrete Ranked Probability Scores for the testing period fc_ar2 <- forecast(mod_ar2) @@ -124,10 +126,12 @@ sum(score_rw$series_1$score) # for estimating model parameters lfo_ar2 <- lfo_cv(mod_ar2, min_t = 40, - fc_horizon = 3) + fc_horizon = 3, + silent = 2) lfo_rw <- lfo_cv(mod_rw, min_t = 40, - fc_horizon = 3) + fc_horizon = 3, + silent = 2) # Plot Pareto-K values and ELPD estimates plot(lfo_ar2) diff --git a/man/loo.mvgam.Rd b/man/loo.mvgam.Rd index 34b184fc..8899fcc5 100644 --- a/man/loo.mvgam.Rd +++ b/man/loo.mvgam.Rd @@ -72,7 +72,8 @@ mod1 <- mvgam(y ~ s(season, bs = 'cc', k = 6), data = rbind(simdat$data_train, simdat$data_test), family = gaussian(), - chains = 2) + chains = 2, + silent = 2) # Inspect the model and calculate LOO conditional_effects(mod1) @@ -85,14 +86,16 @@ mod2 <- update(mod1, formula = y ~ s(season, bs = 'cc', k = 6) + s(season, series, bs = 'fs', xt = list(bs = 'cc'), k = 4), - chains = 2) + chains = 2, + silent = 2) conditional_effects(mod2) loo(mod2) # Now add AR1 dynamic errors to mod2 mod3 <- update(mod2, trend_model = AR(), - chains = 2) + chains = 2, + silent = 2) conditional_effects(mod3) plot(mod3, type = 'trend') loo(mod3) diff --git a/man/mvgam_marginaleffects.Rd b/man/mvgam_marginaleffects.Rd index 16e6648e..5564cfdb 100644 --- a/man/mvgam_marginaleffects.Rd +++ b/man/mvgam_marginaleffects.Rd @@ -86,8 +86,6 @@ arguments.} \item \code{newdata = datagrid(cyl = c(4, 6))}: \code{cyl} variable equal to 4 and 6 and other regressors fixed at their means or modes. \item See the Examples section and the \code{\link[marginaleffects:datagrid]{datagrid()}} documentation. } -\item \code{\link[=subset]{subset()}} call with a single argument to select a subset of the dataset used to fit the model, ex: \code{newdata = subset(treatment == 1)} -\item \code{\link[dplyr:filter]{dplyr::filter()}} call with a single argument to select a subset of the dataset used to fit the model, ex: \code{newdata = filter(treatment == 1)} \item string: \itemize{ \item "mean": Marginal Effects at the Mean. Slopes when each predictor is held at its mean or mode. diff --git a/src/RcppExports.o b/src/RcppExports.o index 03b01119..5cf5ae21 100644 Binary files a/src/RcppExports.o and b/src/RcppExports.o differ diff --git a/src/trend_funs.o b/src/trend_funs.o index 1c0d5a8f..e09f58e3 100644 Binary files a/src/trend_funs.o and b/src/trend_funs.o differ