The goal of expstudy is to provide a set of tools to quickly conduct analysis of an experience study. Commonly used techniques (such as actual-to-expected analysis) are generalized and streamlined so that repetitive coding is avoided.
# This package has not yet been submitted to CRAN, however,
# you can install the development version from GitHub:
# install.packages('devtools')
devtools::install_github('cb12991/expstudy')library(expstudy)
#>
#> Attaching package: 'expstudy'
#> The following objects are masked from 'package:stats':
#>
#> aggregate, filter
#> The following objects are masked from 'package:base':
#>
#> intersect, setdiff, unionFirst you need a dataset that can be used for an experience study. This package provides a sample mortality experience study to aid with examples:
glimpse(mortexp)
#> Rows: 175,491
#> Columns: 18
#> $ AS_OF_DATE <date> 1998-04-30, 1998-05-31, 1998-06-30, 1998-07-31,~
#> $ POLICY_HOLDER <fct> PH_0001, PH_0001, PH_0001, PH_0001, PH_0001, PH_~
#> $ GENDER <fct> MALE, MALE, MALE, MALE, MALE, MALE, MALE, MALE, ~
#> $ SMOKING_STATUS <fct> NON-SMOKER, NON-SMOKER, NON-SMOKER, NON-SMOKER, ~
#> $ UNDERWRITING_CLASS <fct> STANDARD, STANDARD, STANDARD, STANDARD, STANDARD~
#> $ INSURED_DOB <date> 1948-09-10, 1948-09-10, 1948-09-10, 1948-09-10,~
#> $ ISSUE_DATE <date> 1998-04-12, 1998-04-12, 1998-04-12, 1998-04-12,~
#> $ ISSUE_AGE <dbl> 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, ~
#> $ ATTAINED_AGE <dbl> 50, 50, 50, 50, 50, 51, 51, 51, 51, 51, 51, 51, ~
#> $ DURATION_MONTH <int> 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 1~
#> $ DURATION_YEAR <dbl> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, ~
#> $ POLICY_STATUS <fct> DEATH, DEATH, DEATH, DEATH, DEATH, DEATH, DEATH,~
#> $ TERMINATION_DATE <date> 2012-04-13, 2012-04-13, 2012-04-13, 2012-04-13,~
#> $ EXPOSURE <dbl> 0.04931507, 0.08219178, 0.07945205, 0.08219178, ~
#> $ ACTUAL_DEATHS <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ~
#> $ EXPECTED_MORTALITY_RT <dbl> 0.01428571, 0.01428571, 0.01428571, 0.01428571, ~
#> $ EXPECTED_DEATHS <dbl> 0.000704501, 0.001174168, 0.001135029, 0.0011741~
#> $ VARIANCE_DEATHS <dbl> 0.0007040047, 0.0011727896, 0.0011337411, 0.0011~Now you can convert to an expstudy object:
# Only need to specify which variables correlate to metric variable `actuals`,
# `expecteds`, and `exposures`. `variances` and `keys` are beneficial, but not
# required.
es <- expstudy(
data = mortexp,
actuals = ACTUAL_DEATHS,
expecteds = EXPECTED_DEATHS,
exposures = EXPOSURE,
variances = VARIANCE_DEATHS
)With this expstudy object, a quick example can be shown using compile_results():
# We will supply only the dataset and grouping variables to use for the
# summaries to generate a more robust list of results.
results <- compile_results(
expstudy = es,
GENDER,
ATTAINED_AGE
)The compile_results() function combines many expstudy functions to produce many summaries. Since nothing other than the potential groupings were provided, coompile_results() will generate a nested list of summaries:
glimpse(results, give.attr = FALSE)
#> List of 2
#> $ UNFORMATTED:List of 2
#> ..$ METRICS :List of 4
#> .. ..$ AGGREGATE : tbl_es [1 x 7] (S3: tbl_es/tbl_df/tbl/data.frame)
#> .. ..$ BY GENDER : tbl_es [2 x 8] (S3: tbl_es/tbl_df/tbl/data.frame)
#> .. ..$ BY ATTAINED_AGE : tbl_es [83 x 8] (S3: tbl_es/tbl_df/tbl/data.frame)
#> .. ..$ BY GENDER AND ATTAINED_AGE: tbl_es [166 x 9] (S3: tbl_es/tbl_df/tbl/data.frame)
#> ..$ PROPORTIONS:List of 4
#> .. ..$ AGGREGATE : tbl_es [1 x 8] (S3: tbl_es/tbl_df/tbl/data.frame)
#> .. ..$ BY GENDER : tbl_es [2 x 9] (S3: tbl_es/tbl_df/tbl/data.frame)
#> .. ..$ BY ATTAINED_AGE : tbl_es [83 x 9] (S3: tbl_es/tbl_df/tbl/data.frame)
#> .. ..$ BY GENDER AND ATTAINED_AGE: tbl_es [166 x 10] (S3: tbl_es/tbl_df/tbl/data.frame)
#> $ FORMATTED :List of 2
#> ..$ METRICS :List of 4
#> .. ..$ AGGREGATE : tbl_es [1 x 7] (S3: tbl_es/tbl_df/tbl/data.frame)
#> .. ..$ BY GENDER : tbl_es [2 x 8] (S3: tbl_es/tbl_df/tbl/data.frame)
#> .. ..$ BY ATTAINED_AGE : tbl_es [83 x 8] (S3: tbl_es/tbl_df/tbl/data.frame)
#> .. ..$ BY GENDER AND ATTAINED_AGE: tbl_es [166 x 9] (S3: tbl_es/tbl_df/tbl/data.frame)
#> ..$ PROPORTIONS:List of 4
#> .. ..$ AGGREGATE : tbl_es [1 x 8] (S3: tbl_es/tbl_df/tbl/data.frame)
#> .. ..$ BY GENDER : tbl_es [2 x 9] (S3: tbl_es/tbl_df/tbl/data.frame)
#> .. ..$ BY ATTAINED_AGE : tbl_es [83 x 9] (S3: tbl_es/tbl_df/tbl/data.frame)
#> .. ..$ BY GENDER AND ATTAINED_AGE: tbl_es [166 x 10] (S3: tbl_es/tbl_df/tbl/data.frame)An unformatted summary can be used for subsequent analysis…
results$UNFORMATTED$METRICS$`BY GENDER`
#> # A tibble: 2 x 8
#> GENDER ACTUAL_DEATHS EXPECTED_DEATHS EXPOSURE VARIANCE_DEATHS ACTUAL_TO_EXPEC~
#> * <fct> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 FEMALE 134 107. 6098. 107. 1.25
#> 2 MALE 215 145. 8166. 145. 1.48
#> # ... with 2 more variables: ACTUAL_TO_EXPOSED <dbl>, EXPECTED_TO_EXPOSED <dbl>…and a formatted summary can be quickly and simply inserted into a report to distribute nicely.
results$FORMATTED$PROPORTIONS$`BY ATTAINED_AGE`
#> # A tibble: 83 x 9
#> ATTAINED_AGE ACTUAL_DEATHS EXPECTED_DEATHS EXPOSURE VARIANCE_DEATHS
#> * <dbl> <chr> <chr> <chr> <chr>
#> 1 19 0.00 0.21 21.66 0.21
#> 2 20 0.00 0.57 56.52 0.56
#> 3 21 0.00 0.66 65.52 0.66
#> 4 22 0.00 0.71 69.88 0.71
#> 5 23 0.00 0.83 80.32 0.83
#> 6 24 0.00 0.88 84.89 0.88
#> 7 25 0.00 0.98 93.39 0.98
#> 8 26 0.00 1.07 100.19 1.06
#> 9 27 2.00 1.20 111.25 1.20
#> 10 28 0.00 1.30 119.39 1.30
#> # ... with 73 more rows, and 4 more variables: PROP_ACTUAL_DEATHS <chr>,
#> # PROP_EXPECTED_DEATHS <chr>, PROP_EXPOSURE <chr>, PROP_VARIANCE_DEATHS <chr>You can review and trim the summaries down to what you want without much effort. You also can access the formulae that are used in compile_results() to fully customize the analysis to fit your needs.
Please note that the expstudy project is released with a Contributor Code of Conduct. By contributing to this project, you agree to abide by its terms.