scaleLogToOrig: Scale standard deviation between log and original scale.

In lognorm: Functions for the Lognormal Distribution

Description

When comparing values at log scale that have different sd at original scale, better compare log(mean) instead of mu.

Usage

scaleLogToOrig(logmean, sigma)

scaleOrigToLog(mean, sd)

Arguments

logmean	log of the expected value
sigma	standard deviation at log scale
mean	expected value at original scale
sd	standard deviation at original scale

Value

numeric matrix with columns mean, and sd at original scale

Functions

• scaleLogToOrig: get logmean and sigma at log scale

• scaleOrigToLog: get mean and sd at original scale

Examples

  xLog <- data.frame(logmean = c(0.8, 0.8), sigma = c(0.2, 0.3))
  xOrig <- as.data.frame(scaleLogToOrig(xLog$logmean, xLog$sigma))
  xLog2 <- as.data.frame(scaleOrigToLog(xOrig$mean, xOrig$sd))
  all.equal(xLog, xLog2)
  xLog3 <- as.data.frame(getParmsLognormForMoments(xOrig$mean, xOrig$sd^2))
  all.equal(xLog$sigma, xLog3$sigma) # but mu  < logmean 

lognorm documentation built on Nov. 22, 2021, 1:07 a.m.