km

Author: ehsanx

.Rhistory

@@ -1,42 +1,3 @@
-for (var in table_vars) {
-var_formula <- stats::as.formula(paste0("~", var))
-strata_formula <- stats::as.formula(paste0("~", strata_var))
-var_header_row <- as.data.frame(setNames(as.list(c(var, rep("", ncol(header_row) - 1))), names(header_row)))
-results_list[[length(results_list) + 1]] <- var_header_row
-if (is.factor(df[[var]])) {
-df[[var]] <- droplevels(df[[var]])
-unweighted_counts_overall <- table(df[[var]])
-unweighted_pcts_overall <- stats::prop.table(unweighted_counts_overall) * 100
-unweighted_counts_strata <- table(df[[var]], df[[strata_var]])
-unweighted_pcts_strata <- stats::prop.table(unweighted_counts_strata, margin = 2) * 100
-if (mode %in% c("weighted", "mixed")) {
-weighted_counts_overall <- svytable(var_formula, design)
-weighted_pcts_overall <- svymean(var_formula, design, na.rm = TRUE) * 100
-weighted_counts_strata <- svytable(stats::as.formula(paste0("~", var, "+", strata_var)), design)
-weighted_pcts_strata <- svyby(var_formula, strata_formula, design, svymean, na.rm = TRUE)
-}
-for (lvl in levels(df[[var]])) {
-row_data <- data.frame(Variable = "", Level = lvl, stringsAsFactors = FALSE)
-if (mode == "mixed") {
-val <- sprintf("%s (%.1f%%)", format_num(unweighted_counts_overall[lvl], FALSE),
-weighted_pcts_overall[paste0(var, lvl)])
-} else if (mode == "weighted") {
-val <- sprintf("%s (%.1f%%)", format_num(weighted_counts_overall[lvl], TRUE),
-weighted_pcts_overall[paste0(var, lvl)])
-} else {
-val <- sprintf("%s (%.1f%%)", format_num(unweighted_counts_overall[lvl], FALSE),
-unweighted_pcts_overall[lvl])
-}
-row_data$Overall <- val
-for (s_lvl in strata_levels) {
-if (mode == "mixed") {
-val <- sprintf("%s (%.1f%%)", format_num(unweighted_counts_strata[lvl, s_lvl], FALSE),
-weighted_pcts_strata[s_lvl, paste0(var, lvl)] * 100)
-} else if (mode == "weighted") {
-val <- sprintf("%s (%.1f%%)", format_num(unweighted_counts_strata[lvl, s_lvl], TRUE),
-weighted_pcts_strata[s_lvl, paste0(var, lvl)] * 100)
-} else {
-val <- sprintf("%s (%.1f%%)", format_num(unweighted_counts_strata[lvl, s_lvl], FALSE),
 unweighted_pcts_strata[lvl, s_lvl])
 }
 row_data[[s_lvl]] <- val
@@ -510,3 +471,42 @@ rhub::rhub_doctor()
 rhub::rhub_check()
 rhub::rhub_check()
 devtools::check_win_devel()
+# 1. SETUP (do this first, once per session)
+library(devtools)
+library(roxygen2)
+# 3. UPDATE DOCUMENTATION
+devtools::document()  # This runs roxygen2::roxygenise()
+# 3. UPDATE DOCUMENTATION
+devtools::document()  # This runs roxygen2::roxygenise()
+help(package = "utils")
+?tempfile
+# 3. UPDATE DOCUMENTATION
+devtools::document()  # This runs roxygen2::roxygenise()
+# 4. REMOVE OLD VERSION (if testing locally)
+remove.packages("svyTable1")
+# 5. TEST YOUR CHANGES
+devtools::test()  # Run unit tests
+# 6. INSTALL LOCALLY TO TEST
+devtools::install()  # Install and load to test interactively
+# 7. COMPREHENSIVE PACKAGE CHECK
+devtools::check()  # Full check with vignettes
+tools::showNonASCIIfile("R/reportint.R")
+# 7. COMPREHENSIVE PACKAGE CHECK
+devtools::check()  # Full check with vignettes
+# 1. SETUP (do this first, once per session)
+library(devtools)
+library(roxygen2)
+# 3. UPDATE DOCUMENTATION
+devtools::document()  # This runs roxygen2::roxygenise()
+# 4. REMOVE OLD VERSION (if testing locally)
+remove.packages("svyTable1")
+# 5. TEST YOUR CHANGES
+devtools::test()  # Run unit tests
+# 6. INSTALL LOCALLY TO TEST
+devtools::install()  # Install and load to test interactively
+# 7. COMPREHENSIVE PACKAGE CHECK
+devtools::check()  # Full check with vignettes
+# 7. COMPREHENSIVE PACKAGE CHECK
+devtools::check()  # Full check with vignettes
+# 8. IF VIGNETTES ARE IMPORTANT:
+devtools::build_vignettes()

DESCRIPTION

@@ -1,6 +1,6 @@
 Package: svyTable1
 Title: Create Survey-Weighted Descriptive Statistics and Diagnostic Tables
-Version: 0.9.0
+Version: 0.10.0
 Authors@R: c(person("Ehsan", "Karim",
                  email = "ehsan.karim@gmail.com",
                  role = c("aut", "cre")),
@@ -10,33 +10,40 @@ Authors@R: c(person("Ehsan", "Karim",
 Description: A tool to create publication-ready tables from complex 
     survey data, including descriptive summaries and multi-panel 
     interaction reports. It also provides a suite of functions to 
-    evaluate survey-weighted regression models, including coefficient 
-    diagnostics, goodness-of-fit tests, and design-correct AUC 
-    calculations.
+    evaluate survey-weighted regression models and survey-weighted 
+    survival plots, including coefficient diagnostics, 
+    goodness-of-fit tests, and design-correct AUC calculations.
 License: MIT + file LICENSE
 Encoding: UTF-8
 RoxygenNote: 7.3.3
 Imports: 
     dplyr,
+    ggplot2,
     kableExtra,
     knitr,
     magrittr,
     msm,
+    patchwork,
     Publish,
+    purrr,
     rlang,
     rmarkdown,
     rstudioapi,
+    scales,
     stats,
     stringr,
+    survival,
     survey,
     tibble,
     tidyr,
     utils,
     WeightedROC
 Suggests: 
-    ggplot2,
     mice,
     NHANES,
     testthat (>= 3.0.0)
 VignetteBuilder: knitr
 Config/testthat/edition: 3
+Depends: 
+    R (>= 3.5)
+LazyData: true

NAMESPACE

@@ -9,6 +9,7 @@ export(reportint)
 export(svyAUC)
 export(svydiag)
 export(svygof)
+export(svykmplot)
 export(svypooled)
 export(svytable1)
 import(stats)
@@ -19,12 +20,33 @@ importFrom(dplyr,"%>%")
 importFrom(dplyr,across)
 importFrom(dplyr,bind_rows)
 importFrom(dplyr,case_when)
+importFrom(dplyr,count)
+importFrom(dplyr,group_by)
 importFrom(dplyr,if_else)
 importFrom(dplyr,left_join)
 importFrom(dplyr,mutate)
 importFrom(dplyr,relocate)
 importFrom(dplyr,rename)
 importFrom(dplyr,select)
+importFrom(dplyr,sym)
+importFrom(dplyr,ungroup)
+importFrom(ggplot2,aes)
+importFrom(ggplot2,annotate)
+importFrom(ggplot2,coord_cartesian)
+importFrom(ggplot2,element_blank)
+importFrom(ggplot2,element_text)
+importFrom(ggplot2,geom_point)
+importFrom(ggplot2,geom_ribbon)
+importFrom(ggplot2,geom_step)
+importFrom(ggplot2,geom_text)
+importFrom(ggplot2,ggplot)
+importFrom(ggplot2,labs)
+importFrom(ggplot2,scale_color_manual)
+importFrom(ggplot2,scale_fill_manual)
+importFrom(ggplot2,scale_x_continuous)
+importFrom(ggplot2,scale_y_discrete)
+importFrom(ggplot2,theme)
+importFrom(ggplot2,theme_classic)
 importFrom(graphics,abline)
 importFrom(graphics,plot)
 importFrom(graphics,title)
@@ -34,12 +56,15 @@ importFrom(kableExtra,pack_rows)
 importFrom(knitr,kable)
 importFrom(magrittr,"%>%")
 importFrom(msm,deltamethod)
+importFrom(patchwork,plot_layout)
+importFrom(purrr,map_dfr)
 importFrom(rlang,":=")
 importFrom(rlang,.data)
 importFrom(rlang,sym)
 importFrom(rmarkdown,render)
 importFrom(rstudioapi,isAvailable)
 importFrom(rstudioapi,viewer)
+importFrom(scales,comma)
 importFrom(stats,coef)
 importFrom(stats,confint)
 importFrom(stats,fitted)
@@ -64,12 +89,16 @@ importFrom(survey,svyby)
 importFrom(survey,svyciprop)
 importFrom(survey,svydesign)
 importFrom(survey,svyglm)
+importFrom(survey,svykm)
+importFrom(survey,svylogrank)
 importFrom(survey,svymean)
 importFrom(survey,svytable)
 importFrom(survey,svyvar)
 importFrom(survey,withReplicates)
+importFrom(survival,Surv)
 importFrom(tibble,as_tibble)
 importFrom(tibble,tibble)
 importFrom(tidyr,pivot_longer)
 importFrom(tidyr,pivot_wider)
+importFrom(tools,toTitleCase)
 importFrom(utils,browseURL)

R/data.R

@@ -0,0 +1,36 @@
+#' Processed NHANES Mortality Data for Diabetic Adults
+#'
+#' A subset of the NHANES 1999-2010 data, filtered for adults (>=18)
+#' with diabetes and plausible caloric intake (500-3500 kcal).
+#' This data is prepared for survival analysis on the relationship
+#' between caffeine consumption and mortality.
+#'
+#' @format A data frame with 3780 rows and 23 variables:
+#' \describe{
+#'   \item{id}{Respondent sequence/ID number}
+#'   \item{survey.weight}{Full sample 2-year weights}
+#'   \item{psu}{Masked pseudo-PSU}
+#'   \item{strata}{Masked pseudo-stratum}
+#'   \item{caff}{Caffeine consumption (Factor: "No consumption", "<100 mg/day", etc.)}
+#'   \item{stime}{Follow-up time in months}
+#'   \item{status}{Mortality status (0=Censored, 1=Deceased)}
+#'   \item{sex}{Sex (Factor: "Male", "Female")}
+#'   \item{age}{Age in years at interview}
+#'   \item{race}{Race/ethnicity (Factor)}
+#'   \item{smoking}{Smoking status (Factor)}
+#'   \item{bmi.cat}{BMI category (Factor)}
+#'   \item{education}{Education level (Factor)}
+#'   \item{carbohyd}{Carbohydrate in gm}
+#'   \item{alcohol}{Alcohol consumption (Factor)}
+#'   \item{htn}{Hypertension status (Factor)}
+#'   \item{macrovascular}{Macrovascular complications (Character: "No", "Yes")}
+#'   \item{insulin}{Insulin use (Factor: "No", "Yes")}
+#'   \item{survey.cycle}{Survey cycle (Character: "1999-00", etc.)}
+#'   \item{physical.activity}{Physical activity level (Factor)}
+#'   \item{diabetes}{Diabetes status (Factor, all "Yes")}
+#'   \item{cal.total}{Total calories in kcal}
+#'   \item{survey_weight}{Adjusted 6-cycle survey weight (survey.weight / 6)}
+#' }
+#' @source Data originally from NHANES and processed according to the
+#' "Survival analysis: NHANES" tutorial.
+"nhanes_mortality"

R/globals.R

@@ -1,4 +1 @@
-utils::globalVariables(c(
-  ".", "term", "estimate", "conf.low", "conf.high", "group",
-  "Characteristic", "Estimate_CI", "p_value_formatted"
-))
+utils::globalVariables(c("."))

R/svydiag.R

@@ -43,6 +43,7 @@
 #' @importFrom dplyr mutate select
 #' @importFrom tibble tibble
 #' @importFrom stats confint coef vcov
+#' @importFrom rlang .data
 #'
 #' @export
 #'
@@ -122,23 +123,23 @@ svydiag <- function(fit, p_threshold = 0.05, rse_threshold = 30) {
   # 3. Calculate derived metrics, add flags, and finalize the output
   reliability_df <- reliability_df %>%
     dplyr::mutate(
-      RSE_percent = (SE / abs(Estimate)) * 100,
-      CI_Width = CI_Upper - CI_Lower,
-      is_significant = p.value < p_threshold,
-      is_rse_high = RSE_percent >= rse_threshold
+      RSE_percent = (.data$SE / abs(.data$Estimate)) * 100,
+      CI_Width = .data$CI_Upper - .data$CI_Lower,
+      is_significant = .data$p.value < p_threshold,
+      is_rse_high = .data$RSE_percent >= rse_threshold
     ) %>%
     # Reorder and select the final columns for a clean output
     dplyr::select(
-      Term,
-      Estimate,
-      SE,
-      p.value,
-      is_significant,
-      CI_Lower,
-      CI_Upper,
-      CI_Width,
-      RSE_percent,
-      is_rse_high
+      .data$Term,
+      .data$Estimate,
+      .data$SE,
+      .data$p.value,
+      .data$is_significant,
+      .data$CI_Lower,
+      .data$CI_Upper,
+      .data$CI_Width,
+      .data$RSE_percent,
+      .data$is_rse_high
     )
 
   return(reliability_df)