Density distributions of lognormal distributions (lines) get closer to normal density shaded area) as multiplicative standard deviation \(\sigma^*\) decreases down to 1.2 for same \(\mu^* = 1\).
Are already provided with the base stats package. See ?dlnorm
.
getLognormMode(mu = 0.6,sigma = 0.5)
## [1] 1.419068
getLognormMedian(mu = 0.6,sigma = 0.5)
## [1] 1.822119
<- getLognormMoments(mu = 0.6,sigma = 0.5)) (theta
## mean var cv
## [1,] 2.064731 1.210833 0.5329404
Mode < Median < Mean for the right-skewed distribution.
The return type of getLognormMoments
is a matrix.
<- cbind(mean = c(1,1), var = c(0.2, 0.3)^2 )
moments <- getParmsLognormForMoments( moments[,1], moments[,2])) (theta
## mu sigma
## [1,] -0.01961036 0.1980422
## [2,] -0.04308885 0.2935604
The larger the spread, the more skewed is the distribution, here both with an expected value of one.