In FredHasselman/nlRtsa: nonlineaR time seRies analysis.

require(knitr)
require(formatR)
require(dplyr)
options(width=200)
knitr::opts_chunk$set(cache=FALSE,prompt=FALSE,comment=">",message=FALSE,echo=TRUE,warning=FALSE,tidy=TRUE,strip.white=TRUE,size="small", fig.align = "center",fig.show='hold')

Phase Space Reconstruction: Quantifying Recurrent Patterns

You can use R or Matlab to run RQA analyses. This document is based on R. Yoiu can find the assignments for Matlab on Blackboard.

Using R: Install package `nlRtsa`

First, download an install package nlRtsa (under construction)

IMPORTANT!: Make sure you install the package to a library in directory in which you have write access. In the computer rooms at the university, the easiest way to make sure things run smoothly is to first use the RStudio menu on the Packages pane to install a package, e.g., devtools. RStudio will give you the option to install the package in your personal user directory.

require(devtools)
install_github("FredHasselman/nlRtsa")
library(nlRtsa)

If this does not work, run the following code:

require(devtools)
source_url("https://raw.githubusercontent.com/FredHasselman/nlRtsa/master/filesource/nlRtsa_SOURCE.R")

This will source all the functions in nlRtsa into your environment.
Drawback is that you cannot access the manual pages.

Assignment: Phase Space Reconstruction

We'll package fractal to reconstruct some phase spaces.

Install and load these packages by running: in.IT('fractal','rgl')
- Run the folowing code: plot3d(lorenz)
Package fractal includes the 3 dimensions of the Lorenz system in the chaotic regime. We'll reconstruct it based on just dimension X
Use lx <- lorenz[1:2048,1] to reconstruct the phase space. + Find an optimal embedding lag using timeLag, use method = "mutual". + Find the embeddingdimension, using FNN + Plot the phase space (use open3d() to open an interactive plot if you like.