confIntProportion {biostatUZH} | R Documentation |
Compute a confidence interval for a binomial proportion using several
asymptotic and exact methods. The individual methods are also
available as separate functions wald
, wilson
,
agresti
, jeffreys
, and clopperPearson
.
confIntProportion(x, n, conf.level = 0.95)
x |
Number of successes. |
n |
Total number of trials. |
conf.level |
Confidence level for confidence interval. |
A list with the entries:
p |
Estimated proportion. |
CIs |
Dataframe containing the estimated confidence intervals. |
Kaspar Rufibach and Leonhard Held
All the intervals provided in these functions are compared in:
Brown, L.D., Cai, T.T., DasGupta, A. (2001). Interval Estimation for a Binomial Proportion. Statistical Science, 16(2), 101–133.
Functions for some of the intervals provided here are available in Hmisc (see the examples).
## Calculate confidence bounds for a binomial parameter by different methods. x <- 50 n <- 100 ci <- confIntProportion(x, n)$CIs ci plot(0, 0, type = 'n', ylim = c(0, 7), xlim = c(0, 1), xlab = 'p', ylab = '', yaxt = 'n') lines(ci[1, 2:3], c(1, 1)) lines(ci[2, 2:3], c(2, 2)) lines(ci[3, 2:3], c(3, 3)) lines(ci[4, 2:3], c(4, 4)) lines(ci[5, 2:3], c(5, 5)) text(0.5, 0.85, 'wald') text(0.5, 1.85, 'wilson') text(0.5, 2.85, 'agresti') text(0.5, 3.85, 'jeffreys') text(0.5, 4.85, 'clopper') ## compare intervals to those received by the function binconf in Hmisc: if (require("Hmisc")) { binconf(x, n, method = "asymptotic") # Wald binconf(x, n, method = "wilson") # Wilson binconf(x, n, method = "exact") # Clopper-Pearson }