sample.df: Select Random Sampling from Data Frame

In ddavid-evdy/miscR:

Description

Returns a random sample of a given data frame, either simple or stratified, depending on inputs.

Usage

sample.df(df, size, strata = NULL, sample.only = NULL)

Arguments

df	Data frame for selecting stratified random sample.
size	Size of sample to select. If less than 1, a percent will be selected. If greater than 1, an absolute number will be selected.
strata	String, or vector of strings, containing names of fields in df on which to stratify. If NULL, selects a simple random sample.
sample.only	Flag for returning entire dataframe with sample marked, or sample only. If NULL, returns entire dataframe with sample marked.

Details

If strata are specified, and an absolute size is given which is larger than the total number of observations in a given strata cell, no records will be selected from that cell.

Value

Data frame containing selected observations.

Examples

# Return whole dataframe with 10% of observations randomly marked
sample.df(workhrs, 0.10)

# Return randomly selected 12% of the dataframe
sample.df(workhrs, 0.10, sample.only = TRUE)

# Return 21 randomly selected observations
sample.df(workhrs, 21, sample.only = TRUE)

# Return randomly selected 30% of the dataframe, distributed proportionally across marital.status
sample.df(workhrs, 0.30, strata = "marital.status", sample.only = TRUE)

ddavid-evdy/miscR documentation built on May 15, 2019, 1:49 a.m.