This function summarises posterior samples for arbitrary strata. It optionally holds out the observed data (variables that are not ".draw", ".iteration", ".sample", ".chain" ) joins this to the summarised posterior.
Arguments
- samples
A
data.frameof posterior samples with at least a numeric sample variable.- probs
A vector of numeric probabilities to produce quantile summaries for. By default these are the 5%, 20%, 80%, and 95% quantiles which are also the minimum set required for plotting functions to work.
- by
A character vector of variables to summarise by. Defaults to
c("reference_date", ".group").- link_with_obs
Logical, should the observed data be linked to the posterior summary? This is useful for plotting the posterior against the observed data. Defaults to
TRUE.
See also
Functions used for postprocessing of model fits
enw_add_latest_obs_to_nowcast(),
enw_nowcast_samples(),
enw_nowcast_summary(),
enw_posterior(),
enw_pp_summary(),
enw_quantiles_to_long()
Examples
fit <- enw_example("nowcast")
samples <- summary(fit, type = "nowcast_sample")
enw_summarise_samples(samples, probs = c(0.05, 0.5, 0.95))
#> Key: <reference_date, .group>
#> reference_date .group report_date max_confirm location age_group confirm
#> <IDat> <num> <IDat> <int> <fctr> <fctr> <int>
#> 1: 2021-08-03 1 2021-08-22 149 DE 00+ 149
#> 2: 2021-08-04 1 2021-08-22 166 DE 00+ 166
#> 3: 2021-08-05 1 2021-08-22 133 DE 00+ 133
#> 4: 2021-08-06 1 2021-08-22 137 DE 00+ 137
#> 5: 2021-08-07 1 2021-08-22 139 DE 00+ 139
#> 6: 2021-08-08 1 2021-08-22 97 DE 00+ 97
#> 7: 2021-08-09 1 2021-08-22 58 DE 00+ 58
#> 8: 2021-08-10 1 2021-08-22 175 DE 00+ 175
#> 9: 2021-08-11 1 2021-08-22 233 DE 00+ 233
#> 10: 2021-08-12 1 2021-08-22 237 DE 00+ 237
#> 11: 2021-08-13 1 2021-08-22 204 DE 00+ 204
#> 12: 2021-08-14 1 2021-08-22 189 DE 00+ 189
#> 13: 2021-08-15 1 2021-08-22 125 DE 00+ 125
#> 14: 2021-08-16 1 2021-08-22 98 DE 00+ 98
#> 15: 2021-08-17 1 2021-08-22 242 DE 00+ 242
#> 16: 2021-08-18 1 2021-08-22 223 DE 00+ 223
#> 17: 2021-08-19 1 2021-08-22 202 DE 00+ 202
#> 18: 2021-08-20 1 2021-08-22 171 DE 00+ 171
#> 19: 2021-08-21 1 2021-08-22 112 DE 00+ 112
#> 20: 2021-08-22 1 2021-08-22 45 DE 00+ 45
#> reference_date .group report_date max_confirm location age_group confirm
#> cum_prop_reported delay prop_reported mean median sd mad
#> <num> <num> <num> <num> <num> <num> <num>
#> 1: 1 19 0.000000000 149.000 149 0.000000 0.0000
#> 2: 1 18 0.000000000 167.426 167 1.318432 1.4826
#> 3: 1 17 0.000000000 135.789 136 1.854162 1.4826
#> 4: 1 16 0.000000000 141.288 141 2.378805 1.4826
#> 5: 1 15 0.007194245 146.019 146 3.079630 2.9652
#> 6: 1 14 0.000000000 103.853 104 2.969884 2.9652
#> 7: 1 13 0.000000000 62.766 63 2.432110 2.9652
#> 8: 1 12 0.000000000 185.035 185 3.847932 4.4478
#> 9: 1 11 0.000000000 256.031 256 6.498790 5.9304
#> 10: 1 10 0.004219409 267.692 267 7.713536 7.4130
#> 11: 1 9 0.000000000 236.057 235 8.508004 7.4130
#> 12: 1 8 0.015873016 231.407 231 10.083304 10.3782
#> 13: 1 7 0.040000000 164.876 164 9.848127 10.3782
#> 14: 1 6 0.010204082 128.858 128 8.134495 7.4130
#> 15: 1 5 0.012396694 293.309 292 11.938427 11.8608
#> 16: 1 4 0.017937220 293.042 292 15.799489 14.8260
#> 17: 1 3 0.019801980 291.478 290 21.108128 19.2738
#> 18: 1 2 0.070175439 293.178 291 29.440398 28.1694
#> 19: 1 1 0.383928571 308.518 302 51.151522 46.7019
#> 20: 1 0 1.000000000 375.644 363 103.424281 91.1799
#> cum_prop_reported delay prop_reported mean median sd mad
#> q5 q50 q95
#> <num> <num> <num>
#> 1: 149.00 149 149.00
#> 2: 166.00 167 170.00
#> 3: 133.00 136 139.00
#> 4: 138.00 141 146.00
#> 5: 141.00 146 152.00
#> 6: 100.00 104 109.00
#> 7: 59.00 63 67.00
#> 8: 179.00 185 192.00
#> 9: 246.00 256 268.00
#> 10: 256.00 267 282.00
#> 11: 223.00 235 251.00
#> 12: 216.00 231 250.00
#> 13: 150.00 164 181.00
#> 14: 117.00 128 143.00
#> 15: 276.00 292 315.00
#> 16: 268.00 292 320.00
#> 17: 260.00 290 330.00
#> 18: 250.00 291 344.00
#> 19: 239.00 302 399.10
#> 20: 233.95 363 565.05
#> q5 q50 q95