mipfp: GetConfInt – R documentation

Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!

GetConfInt

Computing confidence intervals for the estimated counts and probabilities (deprecated)

Description

This function computes the (asymptotic) Wald confidence intervals at a given significance level for the results generated by Ipfp and ObtainModelEstimates (provided that their option compute.cov was set to TRUE).

Usage

GetConfInt(list.est, alpha = 0.05)

Arguments

`list.est`	A list produced either by `Ipfp` or `ObtainModelEstimates` containing the estimated counts and probabilities as well as their associated standard deviations.
`alpha`	Significance level of the confidence interval corresponding to the 100(1 - α)% confidence level.

Details

The confidence interval of the estimates X.hat, at significance level α is given by

X.hat +/- z(1-α/2) * σ.hat

where σ.hat is the standart deviations of X.hat and z is the inverse of the cumulative distribution function of the standard normal distribution.

Value

A list of matrices containing the upper and lower bounds for the estimated counts and probabilities.

`lower.x`	Lower bounds of the confidence interval for `list.est$x.hat`.
`upper.x`	Upper bounds of the confidence interval for `list.est$x.hat`.
`lower.p`	lower bounds of the confidence interval for `list.est$p.hat`.
`upper.p`	upper bounds of the confidence interval for `list.est$p.hat`.

Warning

Note: this function is deprecated, instead use confint.mipfp.

Author(s)

Johan Barthelemy

Maintainer: Johan Barthelemy johan@uow.edu.au.

References

Smithson, M. (2002). Confidence intervals. Sage Publications.

Examples

# true contingency (2-way) table
true.table <- array(c(43, 44, 9, 4), dim = c(2, 2))
# generation of sample, i.e. the seed to be updated
seed <- ceiling(true.table / 10)
# desired targets (margins)
target.row <- apply(true.table, 2, sum)
target.col <- apply(true.table, 1, sum)
# storing the margins in a list
target.data <- list(target.col, target.row)
# list of dimensions of each marginal constrain
target.list <- list(1, 2)
# calling the Ipfp function
res <- Ipfp(seed, target.list, target.data)
# addint the standart deviations to res (required by GetConfInt)
cov.res <- vcov(res, seed = seed, target.list = target.list, 
                target.data = target.data)
res$p.hat.se <- cov.res$p.hat.se
res$x.hat.se <- cov.res$x.hat.se
# computing and printing the confidence intervals
print(GetConfInt(res))

mipfp

Multidimensional Iterative Proportional Fitting and Alternative Models

v3.2.1

GPL-2

Authors

Johan Barthelemy [aut, cre], Thomas Suesse [aut], Mohammad Namazi-Rad [ctb]

Initial release

2018-08-29