Create an expstudy object

expstudy() creates a subclass, tbl_es, of a dtplyr::lazy_dt(), that stores attributes relating to an experience study.These attributes provide other package functions arguments which reduce time needed to review an assumption.

Usage

expstudy(data, actuals, expecteds, exposures, variances = NULL, keys = NULL)

Arguments

data

the dataset of an experience study. Can be any kind of organized data (e.g., base::data.frame(), tibble::tibble(), etc.) but must be able to be converted to a data.table::data.table().

actuals

the observed variable (or variables) within the experience study (e.g., actual lapse counts for a lapse study).

expecteds

the expected variable (or variables) within the experience study (e.g., actual deaths for a mortality study).

exposures

the exposure variable (or variables) within the experience study. This will be the time the records spans in years (or within (0, 1) for records covering less than a year).

variances

the variance variable (or variables) within the experience study. This may not be available for all studies, thus can be omitted. Variance is used for credibility calculations.

keys

variables that uniquely describe every record within the experience study. This typically is policy number and a measure of time, such as annual or monthly duration. Can be omitted, but providing enables quicker binary searches within the dtplyr::lazy_dt(). For more information, refer to the data.table vignette: vignette('datatable-keys-fast-subset', package = 'data.table').

Value

an expstudy

Examples

  expstudy(
    data = mortexp,
    actuals = ACTUAL_DEATHS,
    expecteds = EXPECTED_DEATHS,
    exposures =  EXPOSURE,
    variances = VARIANCE_DEATHS
  )
#> Source: local data table [175,491 x 18]
#> Call:   `_DT26`
#> 
#>   AS_OF_DATE POLICY_HOLDER GENDER SMOKING_STATUS UNDERWRITING_CLASS INSURED_DOB
#>   <date>     <fct>         <fct>  <fct>          <fct>              <date>     
#> 1 1998-04-30 PH_0001       MALE   NON-SMOKER     STANDARD           1948-09-10 
#> 2 1998-05-31 PH_0001       MALE   NON-SMOKER     STANDARD           1948-09-10 
#> 3 1998-06-30 PH_0001       MALE   NON-SMOKER     STANDARD           1948-09-10 
#> 4 1998-07-31 PH_0001       MALE   NON-SMOKER     STANDARD           1948-09-10 
#> 5 1998-08-31 PH_0001       MALE   NON-SMOKER     STANDARD           1948-09-10 
#> 6 1998-09-30 PH_0001       MALE   NON-SMOKER     STANDARD           1948-09-10 
#> # ℹ 175,485 more rows
#> # ℹ 12 more variables: ISSUE_DATE <date>, ISSUE_AGE <dbl>, ATTAINED_AGE <dbl>,
#> #   DURATION_MONTH <int>, DURATION_YEAR <dbl>, POLICY_STATUS <fct>,
#> #   TERMINATION_DATE <date>, EXPOSURE <dbl>, ACTUAL_DEATHS <dbl>,
#> #   EXPECTED_MORTALITY_RT <dbl>, EXPECTED_DEATHS <dbl>, VARIANCE_DEATHS <dbl>
#> 
#> # Use as.data.table()/as.data.frame()/as_tibble() to access results