ADMUR (version 1.0.3)

relativeRate: Calculates the relative growth (or decline) rate per generation.

Description

Calculates the generational growth/decline rate for a linear piece of a CPL model, or between any two x,y coordinate pairs.

Usage

relativeRate(x, y, generation = 25, N = 1000)

Arguments

x

A numeric vector of length 2, giving the start and end date of linear piece, or a 2 column matrix such that each row is a start and end pair.

y

The corresponding y values such as PDs, or population size (numeric vector of length 2), or a matrix such that each row is a start and end pair.

generation

Years per generation. Default = 25.

N

Number of sections to average the growth rate across.

Value

Returns a numeric vector of values between 0 and 100 representing a 'relative percentage rate per generation'. Negative values indicate a decline through time, positive indicate growth.

Details

The 'relative rate' (growth or decline) of a straight line between two x,y coordinate pairs is the expected generational growth rate across this line. It is calculated always relative to the larger y-value, providing a symmetric measure. E.g., the absolute percentage changes from 80 to 100 to 80 are calculated as +20 The expected rate is the mean average of the conventional rates for N equal sections of the line, as N approaches infinity.

Examples

# NOT RUN {
	x <- c(5600,5500)
	y <- c(75,80)

	# conventional growth/decline rate per 25 yr generation
	100 * exp(log(y[2]/y[1])/((x[1]-x[2])/25)) - 100

	# relative growth/decline rate per 25 yr generation
	relativeRate(x,y)

	x <- c(5600,5500)
	y <- c(480,75)

	# conventional growth/decline rate per 25 yr generation
	100 * exp(log(y[2]/y[1])/((x[1]-x[2])/25)) - 100

	# relative growth/decline rate per 25 yr generation
	relativeRate(x,y)

	x <- c(5600,5500)
	y <- c(480,0)

	# conventional growth/decline rate per 25 yr generation
	100 * exp(log(y[2]/y[1])/((x[1]-x[2])/25)) - 100

	# relative growth/decline rate per 25 yr generation
	relativeRate(x,y)

	# various random rates between 6000 and 5500 BP
	x <- t(matrix(c(6000,5500),2,1000))
	y <- matrix(runif(2000),1000,2)
	conventional <- 100 * exp(log(y[,2]/y[,1])/((x[,1]-x[,2])/25)) - 100
	relative <- relativeRate(x,y)
	plot(relative, conventional)

	
# }