| confint.acs {acs} | R Documentation |
Return upper and lower bounds of given confidence intervals for acs objects.
Description
When passed an acs object, confint will return
a list of two-column dataframes (one dataframe for each variable
specified in parm) including lower and upper bounds for given
confidence intervals. Intervals can be one- or two-sided.
Usage
## S3 method for class 'acs'
confint(object, parm = "all", level = 0.95, alternative = "two.sided", ...)
Arguments
object |
a acs object (or subset). |
parm |
which variables/columns to return confidence intervals for; defaults to "all", which computes confidence intervals for all estimates in the acs object. |
level |
the confidence level required – e.g., .95 = 95% confidence. |
alternative |
whether the interval should be one-sided (i.e., one-tailed – "greater" or "less" – extending to Inf (or -Inf) on one side) or "two-sided". |
... |
additional argument(s) for methods. |
Value
Returns a list of dataframes (one for each variable specified in
parm) of the lower and upper bounds of the confidence interval
for each row of the data.
Author(s)
Ezra Haber Glenn eglenn@mit.edu.
Examples
# load ACS data
data(kansas09)
# confidence intervals for select columns
confint(kansas09[20:25,], parm=c(4,5,10))
# another way to accomplish this
confint(kansas09[20:25,c(4,5,10)])
# store data and extract at will
my.conf <- confint(kansas09)
str(my.conf)
my.conf[32]
my.conf$Universe...TOTAL.POPULATION.IN.THE.UNITED.STATES..U.S..citizen.by.naturalization
# try a different value for level
confint(kansas09[1:10,6], level=.75)
# ... or a one-sided confidence interval
confint(kansas09[1:10,6], level=.75, alternative="greater")
confint(kansas09[1:10,29], level=.75, alternative="less")