josherrickson

summary.lm <- function(object, correlation = FALSE, symbolic.cor = FALSE, ..., vcov. = NULL) {
  ss <- stats::summary.lm(object, correlation, symbolic.cor, ...)

  if (!is.null(vcov.)) {
    if (is.function(vcov.)) {
      vcov. <- vcov.(object, ...)
    }

    ss$cov.unscaled <- vcov./ss$sigma^2
    ss$coefficients[,2] <- sqrt(diag(vcov.))

josherrickson

library(forestplot)

data(mtcars)
mtcars$disp <- mtcars$disp/100

coefs <- summary(mod <- lm(mpg ~ disp + cyl + qsec + wt, data = mtcars))$coeff
cis <- confint(mod)

dat <- data.frame(cbind(coefs[, c(1, 4)], cis))
names(dat) <- c("mean", "pv", "lower", "upper")

josherrickson

suppressMessages(require(devtools))
yesno <- function(...) {
  cat(paste0(..., collapse = ""))
  # For whatever reason, devtools:::yesno returns `TRUE` if you select a No
  # option, and `FALSE` if you select a Yes option
  utils::menu(c("Yes", "No")) != 1
}
utils::assignInNamespace("yesno", yesno, "devtools")
# remove stand-alone `yesno`
rm(yesno)

josherrickson

# This gist has been superceded by the howManyImputations package (https://cran.r-project.org/web/packages/howManyImputations/index.html and https://errickson.net/howManyImputations/).

josherrickson

webuse mheart1s20, clear

* Run the input model, producing the results as well as returning the AIC/BIC
capture program drop modelPlusIC
program modelPlusIC, eclass properties(mi)
   args model
   quiet `model'
   quiet estat ic
   matrix S = r(S)
   * Append the new AIC/BIC to IC

josherrickson

#!/bin/bash

echo "i <- rownames(installed.packages())" > __tmp__.R

grep -oh --include=\*.{Rmd,R} -r * -e "\(library\|require\)([A-Za-z0-9.]\+)" | \
    sed 's_library(_"_'  | \
    sed 's_require(_"_'  | \
    sed 's_)_",_'        | sort | uniq | \
    perl -p -e 's/\n/ /' | \
    sed 's_^_new <- c(_' | \

josherrickson

#!/bin/bash

SEP=","
HEADER=true
STDERR=false

while [ "$#" -gt 0 ]; do
    case "$1" in
        -s) SEP="$2"; shift 2;;
        -c) COL="$2"; shift 2;;

josherrickson

library(dplyr)
library(reshape2)
library(ggplot2)
#' Create a heatmap in ggplot2
#'
#' @param data Data set
#' @param print.cors Default TRUE. Should the correlations be printed in each block of the heatmap?
#' @param reoder Default TRUE. Should the entries be re-ordered via clustering, similar to `heatmap`?
#' @param colors Default c("red", "white", "blue"). Vector of length three defining colors for correaltions (1, 0, -1).
#'