dispunif3: Dispersion statistics for small samples

In thiloklein/dispersion: Dispersion Statistics for Small Samples

Description

This program generates dispersion measures P(S more equal than s) and P(S less equal than s) for data vectors of size n. The approach is to resample from the data vector, choosing 3 elements randomly, and find the defined measures with S drawn from the uniform distribution on the unit 2-simplex, with bss resamples.

Usage

dispunif3(s = c(1, 3, 7), lb = 0.1, ub = 0.9, bss = 2000)

Arguments

s	The observed sample; (not in share vector form) read in from an external file
lb	lower bound for confidence interval
ub	upper bound for confidence interval
bss	number of bootstrap samples

Value

dispunif3 returns a matrix with two columns for dispersion measures P(s >= S) and P(s <= S).

Author(s)

Paul Kattuman, Thilo Klein, Jun Ma

References

Kattuman, P.A. and Ma, J. (2014). Dispersion Statistics for Small Samples. Working Paper, Cambridge Judge Business School.

Examples

## share vector of length 3
dispunif3(s=c(1,3,7),lb=0.1,ub=0.9,bss=2000)

## share vector of length larger 3
dispunif3(s=c(1,2,3,4,5,6,7,8,9,10),lb=0.1,ub=0.9,bss=2000)

thiloklein/dispersion documentation built on May 31, 2019, 10:41 a.m.