Study on Reliable Estimation of Minimum Embedding Dimension Through Statistical Analysis of Nearest Neighbors

This repository contains some of the codes developed during the study of the paper "Reliable Estimation of Minimum Embedding Dimension Through Statistical Analysis of Nearest Neighbors", 2017, by David Chelidze.

📑 Brief description of files:

Here are brief descriptions of the main files/folders used in this project:

• fnn_KennelFraction_Write.m (saves the values of the Kennel Fraction of FNN in text files);
• fnn_KennelFraction_Read.m (reads text files with the values of the Kennel Fraction of FNN);
• fnn_KennelFraction_Plot.m (plots the values of the Kennel Fraction of FNN by the value of the embedded dimension);

Note: files from the "data" folder can be used so that it is not necessary to run the program fnn_KennelFraction_Write.m.

• datasets.m (generates the sets to be analyzed);
• lorenz.m and duffing.m (numerically solve the two systems of differential equations);
• data (contains the data generated by fnn_KennelFraction_Write.m)

💻 Requirements to start this project:

Before starting, make sure you've met the following requirements:

• You have installed MATLAB (https://matlab.mathworks.com/).

🔍 Main references

• Reliable Estimation of Minimum Embedding Dimension Through Statistical Analysis of Nearest Neighbors, 2017, David Chelidze
• Local false nearest neighbors and dynamical dimensions from observed chaotic data, 1993, Abarbanel & Kennel
• https://github.com/iuricode/README-template/blob/main/README-repository/iuricode.md

📝 License

This project is under license. See the LICENSE file for more details.

⬆ Back to the top