poolABC (version 1.0.0)

getmode: Calculate the mode of a distribution

Description

Computes and outputs the mode of the input distribution.

Usage

getmode(x, xlim, weights = NULL, alpha = 0.7, precision = 1000)

Value

a numeric value of the mode of the input distribution.

Arguments

x

is a numeric vector containing the values of the distribution.

xlim

is a vector with two entries.The first entry is the minimum of the x distribution and the second entry is the maximum value of the x distribution. Ideally these values should be the minimum and maximum value of the prior for this particular parameter.

weights

this is an optional input consisting of a vector with the prior weights for the locfit function.

alpha

numeric value with the alpha parameter of the locfit function. The default value is 0.7

precision

value indicating the number of entries evaluated. The larger the value the higher the precision. The default value is 1000.

Details

The locfit::locfit() function is used to fit a local regression to the distribution. The stats::predict() function is then used to predict the y-axis values of the locfit and the mode is defined as the value where that prediction is maximized. Note that if this function is not able to fit a local regression to the distribution, then the mode of the distribution will be assumed to be equal to the median.

Examples

# create a random distribution
x <- rnorm(n = 100, mean = 2, sd = 25)

# compute the mode of the distribution
getmode(x = x, xlim = c(min(x), max(x)))