make.fdt {fdth} | R Documentation |
Frequency distribution table for continuous and categorical variables
Description
Makes a full fdt from a minimal set of information.
Useful to reproduce (when the real data vector is not known) a previous fdt.
Usage
make.fdt(f,
start,
end,
right=FALSE)
make.fdt_cat(f,
categories=NULL,
sort=TRUE,
decreasing=TRUE)
Arguments
f |
a numeric |
start |
the left value of the interval of the first class. |
end |
the last value of the interval of the last class. |
right |
intervals right open (default = |
categories |
the levels of the categorical variable. |
sort |
the levels of the categorical variable will be ordered by frequency. The default is |
decreasing |
if sort argument is |
Details
Given the starting and ending values of the continuous variable table
or the levels of the categorical variable plus the number of intervals and the
absolute frequency values the functions make.fdt
and make.fdt_cat
reconstruct whole fdt or fdt_cat table.
Value
The function make.fdt
returns a list with the slots:
table |
a |
breaks |
a |
The function make.fdt_cat
returns a list whith the slots:
Category |
the levels of the categorical variable. |
f |
absolute frequency, |
rf |
relative frequency, |
rf(%) |
relative frequency in percentages, |
cf |
cumulative frequency; |
cf(%) |
cumulative frequency in percentages, |
Author(s)
Faria, J. C.
Allaman, I. B
Jelihovschi, E. G.
See Also
table
and cut
provided by base package.
Examples
# Numeric
# Making one reference fdt
set.seed(33)
x <- rnorm(1e3,
20,
2)
(ft.r <- fdt(x))
# Making a brand new
(ft.n <- make.fdt(f=ft.r$table$f,
start=13.711,
end=27.229)) # Huumm ..., good, but ... Can it be better?
summary(ft.n,
format=TRUE,
pattern='%.3f') # Is it nice now?
# Categorical
x <- sample(letters[1:5],
1e3,
rep=TRUE)
# Making one reference fdt
(ft.r <- fdt_cat(x))
# Making a brand new
(ft.n <- make.fdt_cat(f=ft.r$f,
categ=ft.r$Category))