R-posts.com

Introduction to mebootSpear() Function

Interested in publishing a one-time post on R-bloggers.com? Press here to learn how.
In the latest version of meboot (v 1.4-8) on CRAN, the function mebootSpear was introduced.  Below is a gentle introduction to its capabilities and a link to a reference paper with further applications to improved Monte Carlo simulations.

The desired properties from the original maximum entropy bootstrap function meboot were retained, while incorporating the additional argument setSpearman.  The original function created bootstrap replicates with unit rank-correlations to the original time-series, setSpearman relaxes this condition.


AirPassengers
The following example will demonstrate the mebootSpear rank-correlation results from the average of 1,000 bootstrap replicates of the AirPassengers dataset.

library(meboot)
output <- mebootSpear(AirPassengers, setSpearman = 0, xmin = 0)$rowAvg

cor(output, AirPassengers, method = "spearman")
[1] 0.01695065



Reference
The following paper is available with additional examples and R-code:

Vinod, Hrishikesh D. and Viole, Fred, Arbitrary Spearman’s Rank Correlations in Maximum Entropy Bootstrap and Improved Monte Carlo Simulations (June 7, 2020). Available at SSRN: 
https://ssrn.com/abstract=3621614 


Exit mobile version