calc_sigma_scaling: Sigma Scaling Calculator

In samplr: Compare Human Performance to Sampling Algorithms

Sigma Scaling Calculator

Description

Calculates the sigma scaling of the chain, and optionally plots the result.

Usage

calc_sigma_scaling(chain, plot = FALSE)

Arguments

chain	Vector of n length, where n is the number of trials or sampler iterations
plot	Boolean. Whether to additionally plot the result.

Details

Sigma scaling is defined as the slope of the regression connecting log time lags and the standard deviation of value changes across time lags. Markets show values of 0.5.

Value

A list containing the vector of possible lags, the sd of the distances at each lag, their log10 counterparts, and the calculated intercept and slope.

Examples

set.seed(1)
chain1 <- sampler_mh(1, "norm", c(0,1), diag(1))
calc_sigma_scaling(chain1[[1]], plot = TRUE)

samplr documentation built on April 4, 2025, 12:30 a.m.