R/day_functions.R

In expstudies: Calculate Exposures, Assign Records to Intervals

#' Create daily exposures
#'
#' This function takes a records file and creates a day for each day the policyholder is active
#'
#' @param records File containing a unique policy key with start and end dates.
#' @param min_date Left truncation date.
#' @param max_date Right truncation date.
#' @examples
#' addDays(records)
#' @export
addDays <- function(records, min_date = NULL, max_date = NULL){
  #Require a unique key.
  if(anyDuplicated(records$key)){
    stop('Key is not unique')
  }

  #Increment start up to min_date
  if(!is.null(min_date)){
    records <- records %>% dplyr::mutate(start = dplyr::if_else(start < min_date, min_date, start))
  }
  #Increment end down to max_date
  if(!is.null(max_date)){
    records <- records %>% dplyr::mutate(end = dplyr::if_else(max_date < end, max_date, end))
  }
  #Filter out values with start dates past end dates, create a days column to assist iteration in makeDays.
  records_mod <- records %>% dplyr::filter(end >= start) %>%
    dplyr::mutate(days = as.integer(end - start + 1))
  #final_size is used to initialize a dataframe of the correct size.
  final_size <- sum(records_mod$days)
  makeDays(as.character(records_mod$key), records_mod$start, records_mod$days, final_size) %>%
    dplyr::mutate(key = as.character(key))
}

#' Estimate the size of daily exposures
#'
#' This function takes a records file and creates a day for each day the policyholder is active
#'
#' @param records File containing a unique policy key with start and end dates.
#' @param min_date Left truncation date.
#' @param max_date Right truncation date.
#' @examples
#' daySize(records)
#' @export
daySize <- function(records, min_date = NULL, max_date = NULL){
  #Increment start up to min_date
  if(!is.null(min_date)){
    records <- records %>% dplyr::mutate(start = dplyr::if_else(start < min_date, min_date, start))
  }
  #Increment end down to max_date
  if(!is.null(max_date)){
    records <- records %>% dplyr::mutate(end = dplyr::if_else(max_date < end, max_date, end))
  }
  #Filter out values with start dates past end dates, calculate total days.
  records %>% dplyr::filter(end >= start) %>%
    dplyr::summarise(num_rows = sum(end - start + 1)) %>%
    unlist()
}