aggregate_data: Measure aggregation by formula

In skolchin/statp: Statistical data analysis support functions

Description

Performs a measures aggregation by formula on a dataset

Usage

aggregate_data(.data, f, id_cols = NULL, value_cols = NULL, na.rm = FALSE)

Arguments

.data	Dataset (data.frame), 'id' column is required.
f	A formula
id_cols	Identification columns except id (string vector or R formula, like ~y)
value_cols	Calculation columns (string vector or R formula, like ~y)
na.rm	TRUE to replace all NAs to zero

Details

The function transforms a dataset so measure IDs become dataframe columns and then applies a formula to calculate aggregated measures.

A formula can be a mesaure ID or any single-parameter function with aggregation operators (+. -, etc). See examples below.

If a dataset contain additional identification variables, id_cols should be provided.

A value_cols parameter specifies which variables are to be aggregated. If omitted, any numeric or logical columns are taken.

Value

Calculation results (data.frame containing id_cols and value_cols) Results are placed to value_cols, id_cols copied without changes

Examples

## Not run: 

d <- data.frame(id = c("i1", "i1", "i2", "i3"), y = c(10, 20, 10, 10), a = c(4,2,3, 1), r = c(4,5,6,7))
aggregate_data(d, f = "log(i1)+i2+i3", id_cols = ~y, value_cols = c("a", "r"), na.rm = TRUE)

d <- data.frame(id = c("i1", "i2", "i3"), a = c(4, 3, 1), r = c(4, 6,7))
aggregate_data(d, f = "i1+i2+ifelse(i3==1,0,-1)", value_cols = c("a", "r"))

## End(Not run)

skolchin/statp documentation built on April 21, 2020, 2:08 a.m.