timestamp function

NAMESPACE

@@ -76,6 +76,8 @@ export(search_for)
 export(single_line_string)
 export(sorted_bars)
 export(splitter)
+export(summary_stats)
+export(ttimestamp)
 export(update_personal_packages)
 export(years_df)
 export(years_floor)

R/dates.R

@@ -1,3 +1,27 @@
+#' Timestamp
+#'
+#' @param time Sys.time() or a date-time object
+#'
+#' @returns A list()
+#'
+#' @examples
+#' timestamp()
+#'
+#' date()
+#'
+#' ttimestamp()
+#'
+#' @autoglobal
+#'
+#' @export
+ttimestamp <- function(time = Sys.time()) {
+  attr(time, "tzone") <- "EST"
+
+  list(
+    strftime(time, "%Y-%m-%d %H:%M:%S"),
+    strftime(time, "%a %b %e %H:%M:%S %Y"))
+}
+
 #' Calculate Number of Days Between Two Dates
 #'
 #' @note This calculation includes the end date in the sum (see example)

R/experiments.R

@@ -318,102 +318,3 @@ is_valid_npi2 <- function(x) {
 
   identical(test, x)
 }
-
-#' Generate Data for Testing
-#'
-#' @param year_seq sequence of years, e.g. `2010:2020`
-#'
-#' @returns A [tibble][tibble::tibble-package] containing the search results.
-#'
-#' @examples
-#' provider_data(2010:2020)
-#'
-#' @autoglobal
-#'
-#' @family mock
-#'
-#' @export
-provider_data <- function(year_seq) {
-
-  lng <- length(year_seq) * 2
-
-  vctrs::vec_rbind(
-    dplyr::tibble(
-      year = {{ year_seq }},
-      group = "A"),
-    dplyr::tibble(
-      year = {{ year_seq }},
-      group = "B")
-  ) |>
-    dplyr::mutate(
-      net_payment = sample(
-        10000:20000, lng
-      )
-    )
-}
-
-#' Generate mock coding/billing data frame
-#'
-#' @param rows number of rows to generate; default is 10
-#'
-#' @param unnest a logical indicating whether to unnest the dates column;
-#'   default is `FALSE`
-#'
-#' @returns A [tibble][tibble::tibble-package]
-#'
-#' @examples
-#' forager_data(rows = 5)
-#'
-#' @autoglobal
-#'
-#' @family mock
-#'
-#' @export
-forager_data <- function(rows = 10, unnest = FALSE){
-
-
-  x <- dplyr::tibble(
-    claim_id        = as.character(wakefield::id(n = rows)),
-    date_of_service = wakefield::date_stamp(
-      n = rows,
-      start         = lubridate::today() - lubridate::dyears(2),
-      random        = TRUE
-    ),
-    payer           = fixtuRes::set_vector(
-      rows,
-      set           = c(
-        "Medicare",
-        "Medicaid",
-        "Cigna",
-        "Humana",
-        "UnitedHealth",
-        "Anthem",
-        "BCBS",
-        "Centene"
-      )
-    ),
-    ins_class       = fixtuRes::set_vector(rows, set = c("Primary", "Secondary")),
-    balance         = as.double(wakefield::income(n = rows, digits = 2) / 300)
-  ) |>
-    dplyr::mutate(
-      date_of_service      = lubridate::as_date(date_of_service),
-      date_of_release      = date_of_service + round(abs(stats::rnorm(
-        length(date_of_service), 11, 4
-      ))),
-      date_of_submission   = date_of_release + round(abs(stats::rnorm(
-        length(date_of_release), 2, 2
-      ))),
-      date_of_acceptance   = date_of_submission + round(abs(stats::rnorm(
-        length(date_of_submission), 3, 2
-      ))),
-      date_of_adjudication = date_of_acceptance + round(abs(stats::rnorm(
-        length(date_of_acceptance), 30, 3
-      )))
-    ) |>
-    tidyr::nest(dates = tidyr::contains("date"))
-
-  if (unnest)
-    x <- tidyr::unnest_wider(x, dates)
-
-  return(x)
-}

R/generate_data.R

@@ -0,0 +1,99 @@
+
+#' Generate Data for Testing
+#'
+#' @param year_seq sequence of years, e.g. `2010:2020`
+#'
+#' @returns A [tibble][tibble::tibble-package] containing the search results.
+#'
+#' @examples
+#' provider_data(2010:2020)
+#'
+#' @autoglobal
+#'
+#' @family mock
+#'
+#' @export
+provider_data <- function(year_seq) {
+
+  lng <- length(year_seq) * 2
+
+  vctrs::vec_rbind(
+    dplyr::tibble(
+      year = {{ year_seq }},
+      group = "A"),
+    dplyr::tibble(
+      year = {{ year_seq }},
+      group = "B")
+  ) |>
+    dplyr::mutate(
+      net_payment = sample(
+        10000:20000, lng
+      )
+    )
+}
+
+#' Generate mock coding/billing data frame
+#'
+#' @param rows number of rows to generate; default is 10
+#'
+#' @param unnest a logical indicating whether to unnest the dates column;
+#'   default is `FALSE`
+#'
+#' @returns A [tibble][tibble::tibble-package]
+#'
+#' @examples
+#' forager_data(rows = 5)
+#'
+#' @autoglobal
+#'
+#' @family mock
+#'
+#' @export
+forager_data <- function(rows = 10, unnest = FALSE){
+
+
+  x <- dplyr::tibble(
+    claim_id        = as.character(wakefield::id(n = rows)),
+    date_of_service = wakefield::date_stamp(
+      n = rows,
+      start         = lubridate::today() - lubridate::dyears(2),
+      random        = TRUE
+    ),
+    payer           = fixtuRes::set_vector(
+      rows,
+      set           = c(
+        "Medicare",
+        "Medicaid",
+        "Cigna",
+        "Humana",
+        "UnitedHealth",
+        "Anthem",
+        "BCBS",
+        "Centene"
+      )
+    ),
+    ins_class       = fixtuRes::set_vector(rows, set = c("Primary", "Secondary")),
+    balance         = as.double(wakefield::income(n = rows, digits = 2) / 300)
+  ) |>
+    dplyr::mutate(
+      date_of_service      = lubridate::as_date(date_of_service),
+      date_of_release      = date_of_service + round(abs(stats::rnorm(
+        length(date_of_service), 11, 4
+      ))),
+      date_of_submission   = date_of_release + round(abs(stats::rnorm(
+        length(date_of_release), 2, 2
+      ))),
+      date_of_acceptance   = date_of_submission + round(abs(stats::rnorm(
+        length(date_of_submission), 3, 2
+      ))),
+      date_of_adjudication = date_of_acceptance + round(abs(stats::rnorm(
+        length(date_of_acceptance), 30, 3
+      )))
+    ) |>
+    tidyr::nest(dates = tidyr::contains("date"))
+
+  if (unnest)
+    x <- tidyr::unnest_wider(x, dates)
+
+  return(x)
+}

R/tidyverse.R

@@ -172,7 +172,7 @@ display_long <- function(data) {
 #'
 #' @param sep separator between combined row data, default is `"-"`
 #'
-#' @returns A `<data.frame>` or `<tibble>` with combined columns
+#' @returns A `<tibble>` with combined columns
 #'
 #' @examples
 #' x <- fuimus:::forager_data()[-5]
@@ -309,3 +309,57 @@ count_wide <- function(df, rows, cols) {
       values_fill = 0
     )
 }
+
+#' Summary statistics
+#'
+#' @param df data frame
+#'
+#' @param condition filter condition, i.e. `patient == "new"`
+#'
+#' @param group_vars variables to group by, i.e. `c(specialty, state, hcpcs, cost)`
+#'
+#' @param summary_vars variables to summarise, i.e. `c(min, max, mode, range)`
+#'
+#' @param arr column to arrange data by, i.e. `cost`
+#'
+#' @param digits Number of digits to round to, default is 3
+#'
+#' @returns A `<tibble>` with the summarized data
+#'
+#' @examples
+#' x <- dplyr::tibble(
+#'    provider = sample(c("A", "B", "C"), size = 200, replace = TRUE),
+#'    city = sample(c("ATL", "NYC"), size = 200, replace = TRUE),
+#'    charges = sample(1000:2000, size = 200),
+#'    payment = sample(1000:2000, size = 200))
+#'
+#' summary_stats(
+#'    x,
+#'    condition    = city == "ATL",
+#'    group_vars   = provider,
+#'    summary_vars = c(charges, payment),
+#'    arr          = provider)
+#' @autoglobal
+#'
+#' @export
+summary_stats <- function(df,
+                          condition = NULL,
+                          group_vars = NULL,
+                          summary_vars = NULL,
+                          arr = NULL,
+                          digits = 3) {
+
+  df |>
+    dplyr::filter({{ condition }}) |>
+    dplyr::summarise(
+      dplyr::across({{ summary_vars }},
+                    list(median = \(x) stats::median(x, na.rm = TRUE),
+                         mean = \(x) mean(x, na.rm = TRUE),
+                         sd = \(x) stats::sd(x, na.rm = TRUE)),
+                    .names = "{.col}_{.fn}"),
+      n = dplyr::n(),
+      .by = ({{ group_vars }})) |>
+    dplyr::arrange(dplyr::desc({{ arr }})) |>
+    dplyr::mutate(dplyr::across(
+      dplyr::where(is.double), ~roundup(., digits = digits)))
+}