View source: R/veg_cover_functions.R
beta_prior_helper | R Documentation |
This function can be used to explore alternative beta distributions to use
with the beta_prior
argument of functions get_cover_quantiles
and get_cover_difference
. It returns parameter values of distributions
corresponding to a prior expected value for stratum vegetation cover and,
optionally, draws a graph of the alternative distributions.
beta_prior_helper(cover, a = c(2, 4, 8, 16), plot = FALSE)
cover |
Expected value (i.e. maximum likelihood value) of vegetation cover. Must be a single value greater than 1.0. |
a |
Candidate values for the first parameter of the beta distribution.
Default is |
plot |
A logical value (default is |
This function interprets the expected vegetation cover value as the mode of
the beta distribution, i.e. the maximum likelihood value. For this reason it
will only accept values greater than 1.0 for the first beta parameter
(argument 'a'
). A beta(1,1) distribution is equivalent to a uniform
distribution of cover values, while beta(a, b)
distributions with
a < 1
are concave, i.e. the mode is undefined.
Given an expected (maximum likelihood) cover value, and candidate values for the first beta parameter, the function calculates corresponding values for the second beta parameter and returns a matrix with columns for the beta parameters (named 'a' and 'b'), and summary statistics.
A matrix with columns for the beta parameters ('a' and 'b') plus
columns for the median cover and the c(0.05, 0.95)
quantiles
(end-points of central 90% interval). The input expected cover value is
attached as the attribute "cover"
.
# Return parameter values and draw a graph of beta distributions for
# an expected stratum cover value of 40\%
params <- beta_prior_helper(0.4, plot=TRUE)
# Retrieve the input expected cover value used for calculations
attr(params, "cover")
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.