prune: Prune a numeric vector
In danielmarcelino/SciencesPo: A Tool Set for Analyzing Political Behavior Data

Description Usage Arguments Details Value Examples

prune returns a pruned numeric vector.

1 2	prune(x, method = c("value", "percentile"), low = NULL, high = NULL, replace = NULL)

`x`	A numeric vector.
`method`	A character string indicating the desired method of pruning with `"value"` being the default. This must be (an abbreviation of) one of the strings `"value"` or `"percentile"`. See Details for more information.
`low`	The lower value/percentile for pruning See Details for more information.
`high`	The upper value/percentile for pruning See Details for more information.
`replace`	Either `NULL` ("rounds" values according to the low and high arguments), `NA`, or a single value that will replace the pruned values. The default is NULL.

prune returns a pruned version of the numeric vector x. NA values in x are ignored during the pruning process but are preserved in the output. prune will do one-sided pruning if only one low or high argument is provided (e.g. prune(x, low=-1) will prune x at a lower value of -1).

prune is designed to be readable from the function call. For example:

prune(x, "value", low=-1, high=1) can be read as "prune x at -1 and 1".
prune(x, "percentile", low=.05, high=.95) can be read as "prune x at the 5th and 95th percentiles".

The arguments low and high are used based on method. prune offers several different options for method:

value: low and high are used as raw values (e.g. .05 is the value .05).
percentile: low and high are used as percentiles (e.g. .05 is 5th percentile).

The output of prune is a pruned numeric vector with the same length as x.

set.seed(51)
x <- rnorm(1e4)
summary(x)

# pruning at the values -1 and 1!
x_val <- prune(x, low=-1, high=1)
summary(x_val)

# pruning at 5th and 95th percentiles!
x_per <- prune(x, "percentile", low=.05, high=.95)
summary(x_per)

# pruning values above the 95th percentile!
x_hi <- prune(x, "percentile", high=.95)
summary(x_hi)

# pruning values converted to NAs!
x_NA <- prune(x, "percentile", low=.05, high=.95, replace=NA)
summary(x_NA)