post_trend: Trend through Posterior Distributions

In rBatt/trawlDiversity: Long-term changes in species richness along the North American coastline

Description

Fit a trend to a time series of posterior estimates. Ends up being similar to bootstrapping.

Usage

post_trend(x, Y, nSamp = 100)

Arguments

x	time step
Y	matrix of posterior samples. Each column is a time step, each row is a sample of the posterior
nSamp	number of times each set of posterior samples should be resampled to fit trend. I.e., number of bootstrap iterations.

Details

Uses Kendall's Tau and accounts for serial correlation in the time series (up to ARMA(2,2)). Uses Kendall::Kendall. Accounts for ties.

Value

A named numeric vector of length 3, containing Kendall's Tau, the associated Z statistic, and the p-value

See Also

tsTau

Examples

n <- 20
x <- 1:n
slope <- 0.5
n_post <- 50
ar_coeff <- 0.85
y <- c(slope*x+arima.sim(model=list(ar=c(ar_coeff)), n=n))
y_post <- sapply(y, function(x)rnorm(n_post, mean=x, sd=1))
post_trend(x, y_post, 10)

rBatt/trawlDiversity documentation built on Aug. 14, 2021, 1:01 p.m.