Given a summarized results dataframe, unnest the summary results column and use the value at risk (VaR) column to identify all the elements that are outliers (having a VaR >= two standard deviations)

identify_outliers(results)

Arguments

results

Scenario summary results

Value

The supplied dataframe with the following additional columns:

  • ale_var_zscore - Annual loss z-score

  • outlier - Logical flag when the z-score is greater than or equal to two

Examples

data(mc_scenario_summary) identify_outliers(mc_scenario_summary)
#> # A tibble: 56 x 20 #> scenario_id domain_id control_descrip… results loss_events_mean #> <chr> <chr> <list> <list> <dbl> #> 1 RS-01 ORG <list [7]> <tibbl… 7.48 #> 2 RS-02 ORG <list [7]> <tibbl… 7.48 #> 3 RS-03 ORG <list [7]> <tibbl… 1.72 #> 4 RS-04 ORG <list [8]> <tibbl… 2.73 #> 5 RS-05 ORG <list [5]> <tibbl… 4.21 #> 6 RS-06 POL <list [4]> <tibbl… 0 #> 7 RS-07 POL <list [4]> <tibbl… 0 #> 8 RS-08 POL <list [4]> <tibbl… 0.042 #> 9 RS-09 COMP <list [2]> <tibbl… 0 #> 10 RS-10 COMP <list [2]> <tibbl… 1.90 #> # … with 46 more rows, and 15 more variables: loss_events_median <dbl>, #> # loss_events_min <dbl>, loss_events_max <dbl>, ale_median <dbl>, #> # ale_max <dbl>, ale_var <dbl>, sle_mean <dbl>, sle_median <dbl>, #> # sle_min <dbl>, sle_max <dbl>, mean_tc_exceedance <dbl>, #> # mean_diff_exceedance <dbl>, mean_vuln <dbl>, ale_var_zscore <dbl>, #> # outlier <lgl>