remix (version 2.1)

remix: Remix and describe.

Description

A quick and easy function for describing datasets.

Usage

remix(formula = cbind(...) ~ ., data = NULL, funs = c(mean, sd, quantile, n, na), ..., cum = FALSE, margin = 0:2, addmargins = FALSE, useNA = c("no", "ifany", "always"), propNA = TRUE, revert = FALSE, method = c("pearson", "kendall", "spearman"), times = NULL, followup = FALSE, test = FALSE, test.summarize = test.summarize.auto, test.survival = test.survival.logrank, test.tabular = test.tabular.auto, show.test = display.test, plim = 4, show.method = TRUE, label = FALSE)

Arguments

formula

a formula (see Details).

data

a data.frame.

funs

functions for describing numeric variable. Can be c(fun1, fun2, fun3) or c("fun1", "fun2", "fun3") or a list.

...

further arguments (all passed to funs), for example na.rm = TRUE

cum

should cumulated frequencies be reported?

margin

index, or vector of indices to generate proportion in frequency tables (0: cell, 1: row, 2: col).

addmargins

whether to add margins

useNA

whether to include NA as a level (factor)

propNA

whether to include NA in proportion calculation

revert

whether to regroup factors or numeric variables when crossing factor with numeric variables

method

a character string indicating which correlation coefficient is to be used. One of "pearson", "kendall", or "spearman", can be abbreviated.

times

vector of times (see ?summary.survival un package survival)

followup

whether to display follow-up time

test

whether to perform tests

test.summarize

a function of two arguments (continuous variable and grouping variable) used to compare continuous variable, that return a list of two components : p.value and method (the test name). See test.summarize.auto, test.summarize.kruskal, test.summarize.oneway.equalvar, or test.summarize.unequalvar for example of such functions. Users can provide their own function.

test.survival

a function of one argument (a formula) used to compare survival estimations, that returns the same components as created by test.summarize. See test.survival.logrank. Users can provide their own function.

test.tabular

a function of three arguments (two categorical variables and a logical na) used to test association between two factors, that returns the same components as created by test.summarize. See test.tabular.auto and test.tabular.fisher. Users can provide their own function.

show.test

a function used to display the test result. See display.test.

plim

number of digits for the p value

show.method

should display the test name?

label

whether to display labels of variables (using label in package Hmisc)

Value

A remix object, basically a list with descriptive tables. It uses ascii package for printing output, and can be use with ascii function.

See Also

cast (reshape) and summary.formula (Hmisc).

Examples

parwidth <- getOption("width")
options(width = 100)

library(remix)
remix(data = iris)
remix(cbind(...) ~ ., iris[, sapply(iris, is.numeric)], funs = c(median, mad, min, max))
remix(cbind(Sepal.Length, I(Sepal.Width^2)) ~ Species, iris, funs = quantile, probs = c(1/3, 2/3))
remix(Sepal.Length + Sepal.Width ~ Petal.Length + Petal.Width, iris)
remix(cbind(Sepal.Length, Sepal.Width) ~ cbind(Petal.Length, Petal.Width), iris)
remix(... ~ ., esoph, cum = TRUE)
remix(alcgp ~ tobgp, esoph, cum = TRUE)
remix(Surv(time, status) ~ x, data = aml, times = seq(0, 120, 12))

options(width = parwidth)