Precision measures

superb comes with a few built-in measures of precisions. All SE.fct() functions produces an interval width; all CI.fct() produces the lower and upper limits of an interval. See htc14,htc15superb for more. "superbPlot-compatible" precision measures must have these parameters:

SE.mean(x)

CI.mean(x, gamma)

SE.median(x)

CI.median(x, gamma)

SE.hmean(x)

CI.hmean(x, gamma)

SE.gmean(x)

CI.gmean(x, gamma)

SE.var(x)

CI.var(x, gamma)

SE.sd(x)

CI.sd(x, gamma)

SE.MAD(x)

CI.MAD(x, gamma)

SE.IQR(x)

CI.IQR(x, gamma)

SE.fisherskew(x)

CI.fisherskew(x, gamma)

SE.pearsonskew(x)

CI.pearsonskew(x, gamma)

SE.fisherkurtosis(x)

CI.fisherkurtosis(x, gamma)

Arguments

x

a vector of numbers, the sample data (mandatory);

gamma

a confidence level for CI (default 0.95).

Value

a measure of precision (SE) or an interval of precision (CI).

References

Examples

# the confidence interval of the mean for default 95% and 90% confidence level
CI.mean( c(1,2,3) )
#> [1] -0.4841377  4.4841377
CI.mean( c(1,2,3), gamma = 0.90)
#> [1] 0.3141455 3.6858545

# Standard errors for standard deviation, for MAD and for fisher skew
SE.sd( c(1,2,3) )
#> [1] 0.5
SE.MAD( c(1,2,3) )
#> [1] 0.4606589
SE.fisherskew( c(1,2,3) )
#> [1] 1.224745