Confidence intervals for generalized means

Computes confidence intervals for three population generalized means (square-root, geometric, and harmonic) using a vector of response scores as input. The square-root mean requires non-negative scores. The geometric and harmonic means require positive scores. The standard errors are recovered from the confidence intervals.

Usage

ci.mean.gen(alpha, y)

Arguments

alpha

alpha level for 1-alpha confidence

y

vector of scores

Value

Returns a 1-row matrix. The columns are:

• Estimate - estimated mean

• SE - recovered standard error

• LL - lower limit of the confidence interval

• UL - upper limit of the confidence interval

Examples

y <- c(32, 47, 28, 15, 20, 41, 87)
ci.mean.gen(.05, y)
#>             Estimate       SE       LL       UL
#> Square-root 35.79395 8.141122 18.64498 58.48619
#> Geometric   33.26410 7.633328 19.47124 56.82741
#> Harmonic    29.07073 7.978149 19.35236 58.39602

# Should return:
#             Estimate       SE       LL       UL
# Square-root 35.79395 8.141122 18.64498 58.48619
# Geometric   33.26410 7.633328 19.47124 56.82741
# Harmonic    29.07073 7.978149 19.35236 58.39602