README.md

In PhilippVWC/myBayes: Functions for time series analysis based on Bayesian methods

Package myBayes

Overview

myBayes is an R package that serves as a collection of probabilistic methods for bayesian inference on dynamical systems with a strong focus on parametric and nonparametric choatic maps. It is essentially a toolbox for parameter estimation, time series filter and prediction. The developer does not make himself responsable for potential errors, code stabililty nor mathematical correctness of the underlying logic.

Contents

Among the most important functions available in myBayes are:

Function | Purpose ---------|---------- MH | Simmulated annealing for optimization in one dimension. Lik_gsm_cpp | A Gaussian Likelihood given time series data of the general symmetric map. blurrVector | For statistical dispersion of deterministic vectors. CeBePa | For approximate parameter transformation from Gaussian distribution to Beta distribution collapse | Routine to remove equal and adjacent values of a vector. myOpt_N | Optimization in more than two dimensions with multiple algorithms (sim. Annealing, BFGS, etc.) axLabels3d | Individual manipulation of axis label annotations for 3d persp plots.

Special remarks

Many routines are implemented in C++ in order to increase performance.

Dependencies

C++ integration is accomplished by Dirk Eddelbuettels Rcpp package (For more information see Rcpp). An installation of Rcpp is crucial for proper functionality of myBayes.

Installation

Install package devtools and install from github repository:

install.packages("devtools")
devtools::install_github("PhilippVWC/myBayes")

Contact

For further remarks, please contact me under ([e-mail]), Philipp van Wickevoort Crommelin.

PhilippVWC/myBayes documentation built on Oct. 2, 2020, 8:25 a.m.