Calculate the geometric standard deviation (GSD)

The GSD is the standard deviation of the geometric mean for of a set of non-negative,non-zero observations.The GSD is a dimensionless (unitless) multiplicative factor. It describes the range from the mean divided by GSD to the mean multiplied by GSD. Because of its multiplicative nature, the GSD cannot be added/subtracted from the mean.

Usage

geom.sd(x, ..., na.rm = FALSE)

Arguments

x: a numeric vector, matrix or data frame
...: additional parameters to be passed to 'geom.mean()'
na.rm: indicate whether or not to remove NAs

Value

a numeric vector of length equal or greater than 1L.

Details

In order to calculate the GSD, we use imputation methods to deal with missingness and preserve the structure of the matrix or data frame. Note that in some cases, this method might lead to an artificial increase or decrease in the association between outcome and original independent variables.

Examples

d <- rlnorm(60,1,1.3)
geom.sd(d)
#> [1] 4.787954

x <- matrix(rlnorm(60,1,1.4), ncol = 4)
mapply(geom.sd, as.data.frame(x))
#>       V1       V2       V3       V4 
#> 5.386534 4.914967 2.687496 4.377672