pareto_qq_test: Q-Q Plot to test for Pareto Distribution

In ptsuite: Tail Index Estimation for Power Law Distributions

Description

This function can be used as a first step to identify whether the data is Pareto distributed before estimating the tail index. If most of the data points appear to be distributed along a line, it is possible that the data may be Pareto. Conversely, if most of the data are distributed non-linearly, then the data is most probably not Pareto distributed.

Usage

pareto_qq_test(dat)

Arguments

dat	Data to be tested for Pareto distribution

Details

This function plots the quantiles of the standard exponential distribution on the x-axis and the log values of the provided data on the y-axis. If Pareto data was supplied, a log transformation of this data would result in an exponential distribution with mean α. These data points would then show up on the QQ-plot as a line with slope 1/α.

The function makes use of the plotly package if available and installed or if not, defaults to the standard R plot.

Value

A Q-Q plot either using plotly if package is available or else a standard R plot.

Examples

x <- generate_pareto(10000, 5, 2)
pareto_qq_test(x)

ptsuite documentation built on May 1, 2019, 10:51 p.m.