This article is based on creating a termination study using sample
data that comes with the actxps package. For information on transaction
studies, see vignette("transactions").
The actxps package includes a data frame containing simulated census data for a theoretical deferred annuity product with an optional guaranteed income rider. The grain of this data is one row per policy.
library(actxps)
library(dplyr)
census_dat
#> # A tibble: 20,000 × 11
#>    pol_num status  issue_date inc_guar qual    age product gender wd_age premium
#>      <int> <fct>   <date>     <lgl>    <lgl> <int> <fct>   <fct>   <int>   <dbl>
#>  1       1 Active  2014-12-17 TRUE     FALSE    56 b       F          77     370
#>  2       2 Surren… 2007-09-24 FALSE    FALSE    71 a       F          71     708
#>  3       3 Active  2012-10-06 FALSE    TRUE     62 b       F          63     466
#>  4       4 Surren… 2005-06-27 TRUE     TRUE     62 c       M          62     485
#>  5       5 Active  2019-11-22 FALSE    FALSE    62 c       F          67     978
#>  6       6 Active  2018-09-01 FALSE    TRUE     77 a       F          77    1288
#>  7       7 Active  2011-07-23 TRUE     TRUE     63 a       M          65    1046
#>  8       8 Active  2005-11-08 TRUE     TRUE     58 a       M          58    1956
#>  9       9 Active  2010-09-19 FALSE    FALSE    53 c       M          64    2165
#> 10      10 Active  2012-05-25 TRUE     FALSE    61 b       M          73     609
#> # ℹ 19,990 more rows
#> # ℹ 1 more variable: term_date <date>The data includes 3 policy statuses: Active, Death, and Surrender.
Let’s assume we’re interested in calculating the probability of surrender over one policy year. We cannot simply calculate the proportion of policies in a surrendered status as this does not represent an annualized surrender rate.
In order to calculate annual surrender rates, we need to break each policy into multiple records. There should be one row per policy per year.
The expose() family of functions is used to perform this
transformation.
exposed_data <- expose(census_dat, end_date = "2019-12-31",
                       target_status = "Surrender")
exposed_data
#> 
#> ── Exposure data ──
#> 
#> • Exposure type: policy_year
#> • Target status: Surrender
#> • Study range: 1900-01-01 to 2019-12-31
#> 
#> # A tibble: 141,252 × 15
#>    pol_num status issue_date inc_guar qual    age product gender wd_age premium
#>      <int> <fct>  <date>     <lgl>    <lgl> <int> <fct>   <fct>   <int>   <dbl>
#>  1       1 Active 2014-12-17 TRUE     FALSE    56 b       F          77     370
#>  2       1 Active 2014-12-17 TRUE     FALSE    56 b       F          77     370
#>  3       1 Active 2014-12-17 TRUE     FALSE    56 b       F          77     370
#>  4       1 Active 2014-12-17 TRUE     FALSE    56 b       F          77     370
#>  5       1 Active 2014-12-17 TRUE     FALSE    56 b       F          77     370
#>  6       1 Active 2014-12-17 TRUE     FALSE    56 b       F          77     370
#>  7       2 Active 2007-09-24 FALSE    FALSE    71 a       F          71     708
#>  8       2 Active 2007-09-24 FALSE    FALSE    71 a       F          71     708
#>  9       2 Active 2007-09-24 FALSE    FALSE    71 a       F          71     708
#> 10       2 Active 2007-09-24 FALSE    FALSE    71 a       F          71     708
#> # ℹ 141,242 more rows
#> # ℹ 5 more variables: term_date <date>, pol_yr <int>, pol_date_yr <date>,
#> #   pol_date_yr_end <date>, exposure <dbl>These functions create exposed_df objects, which are a
type of data frame with some additional attributes related to the
experience study.
Now that the data has been “exposed” by policy year, the observed annual surrender probability can be calculated as:
As a default, the expose() function calculates exposures
by policy year. This can also be accomplished with the function
expose_py(). Other implementations of expose()
include:
expose_cy = exposures by calendar yearexpose_cq = exposures by calendar quarterexpose_cm = exposures by calendar monthexpose_cw = exposures by calendar weekexpose_pq = exposures by policy quarterexpose_pm = exposures by policy monthexpose_pw = exposures by policy weekSee vignette("exposures") for further details on
exposure calculations.
The exp_stats() function creates a summary of observed
experience data. The output of this function is an exp_df
object.
exp_stats(exposed_data)
#> 
#> ── Experience study results ──
#> 
#> • Target status: Surrender
#> • Study range: 1900-01-01 to 2019-12-31
#> 
#> # A tibble: 1 × 4
#>   n_claims claims exposure  q_obs
#>      <int>  <int>    <dbl>  <dbl>
#> 1     2869   2869  132634. 0.0216See vignette("exp_summary") for further details on
exposure calculations.
If the data frame passed into exp_stats() is grouped
using dplyr::group_by(), the resulting output will contain
one record for each unique group.
exp_res <- exposed_data |>
  group_by(pol_yr, inc_guar) |>
  exp_stats()
exp_res
#> 
#> ── Experience study results ──
#> 
#> • Groups: pol_yr and inc_guar
#> • Target status: Surrender
#> • Study range: 1900-01-01 to 2019-12-31
#> 
#> # A tibble: 30 × 6
#>    pol_yr inc_guar n_claims claims exposure   q_obs
#>     <int> <lgl>       <int>  <int>    <dbl>   <dbl>
#>  1      1 FALSE          56     56    7720. 0.00725
#>  2      1 TRUE           46     46   11532. 0.00399
#>  3      2 FALSE          92     92    7103. 0.0130 
#>  4      2 TRUE           68     68   10612. 0.00641
#>  5      3 FALSE          67     67    6447. 0.0104 
#>  6      3 TRUE           57     57    9650. 0.00591
#>  7      4 FALSE         123    123    5799. 0.0212 
#>  8      4 TRUE           45     45    8737. 0.00515
#>  9      5 FALSE          97     97    5106. 0.0190 
#> 10      5 TRUE           67     67    7810. 0.00858
#> # ℹ 20 more rowsTo derive actual-to-expected rates, first attach one or more columns
of expected termination rates to the exposure data. Then, pass these
column names to the expected argument of
exp_stats().
expected_table <- c(seq(0.005, 0.03, length.out = 10), 0.2, 0.15, rep(0.05, 3))
# using 2 different expected termination rates
exposed_data <- exposed_data |>
  mutate(expected_1 = expected_table[pol_yr],
         expected_2 = ifelse(exposed_data$inc_guar, 0.015, 0.03))
exp_res <- exposed_data |>
  group_by(pol_yr, inc_guar) |>
  exp_stats(expected = c("expected_1", "expected_2"))
exp_res
#> 
#> ── Experience study results ──
#> 
#> • Groups: pol_yr and inc_guar
#> • Target status: Surrender
#> • Study range: 1900-01-01 to 2019-12-31
#> • Expected values: expected_1 and expected_2
#> 
#> # A tibble: 30 × 10
#>    pol_yr inc_guar n_claims claims exposure   q_obs expected_1 expected_2
#>     <int> <lgl>       <int>  <int>    <dbl>   <dbl>      <dbl>      <dbl>
#>  1      1 FALSE          56     56    7720. 0.00725    0.005        0.03 
#>  2      1 TRUE           46     46   11532. 0.00399    0.005        0.015
#>  3      2 FALSE          92     92    7103. 0.0130     0.00778      0.03 
#>  4      2 TRUE           68     68   10612. 0.00641    0.00778      0.015
#>  5      3 FALSE          67     67    6447. 0.0104     0.0106       0.03 
#>  6      3 TRUE           57     57    9650. 0.00591    0.0106       0.015
#>  7      4 FALSE         123    123    5799. 0.0212     0.0133       0.03 
#>  8      4 TRUE           45     45    8737. 0.00515    0.0133       0.015
#>  9      5 FALSE          97     97    5106. 0.0190     0.0161       0.03 
#> 10      5 TRUE           67     67    7810. 0.00858    0.0161       0.015
#> # ℹ 20 more rows
#> # ℹ 2 more variables: ae_expected_1 <dbl>, ae_expected_2 <dbl>autoplot() and autotable()The autoplot() and autotable() functions
create visualizations and summary tables. See
vignette("visualizations") for full details on these
functions.
summary()Calling the summary() function on an exp_df
object re-summarizes experience results. This also produces an
exp_df object.
summary(exp_res)
#> 
#> ── Experience study results ──
#> 
#> • Target status: Surrender
#> • Study range: 1900-01-01 to 2019-12-31
#> • Expected values: expected_1 and expected_2
#> 
#> # A tibble: 1 × 8
#>   n_claims claims exposure  q_obs expected_1 expected_2 ae_expected_1
#>      <int>  <int>    <dbl>  <dbl>      <dbl>      <dbl>         <dbl>
#> 1     2869   2869  132634. 0.0216     0.0242     0.0209         0.892
#> # ℹ 1 more variable: ae_expected_2 <dbl>If additional variables are passed to ..., these
variables become groups in the re-summarized exp_df
object.
summary(exp_res, inc_guar)
#> 
#> ── Experience study results ──
#> 
#> • Groups: inc_guar
#> • Target status: Surrender
#> • Study range: 1900-01-01 to 2019-12-31
#> • Expected values: expected_1 and expected_2
#> 
#> # A tibble: 2 × 9
#>   inc_guar n_claims claims exposure  q_obs expected_1 expected_2 ae_expected_1
#>   <lgl>       <int>  <int>    <dbl>  <dbl>      <dbl>      <dbl>         <dbl>
#> 1 FALSE        1601   1601   52123. 0.0307     0.0235      0.03          1.31 
#> 2 TRUE         1268   1268   80511. 0.0157     0.0247      0.015         0.637
#> # ℹ 1 more variable: ae_expected_2 <dbl>