https://github.com/who-else-but-arjun/tvar_elcentro
Analyzing a non-stationary signal, such as the El Centro ground motion data, using a 3rd-order Time-Varying Autoregressive (TVAR) model. The goal is to estimate AR coefficients over time, create a heatmap to visualize their evolution, and provide an analysis of the observed trends and patterns that reflect the signal's non-stationary behavior.
https://github.com/who-else-but-arjun/tvar_elcentro
Last synced: about 2 months ago
JSON representation
Analyzing a non-stationary signal, such as the El Centro ground motion data, using a 3rd-order Time-Varying Autoregressive (TVAR) model. The goal is to estimate AR coefficients over time, create a heatmap to visualize their evolution, and provide an analysis of the observed trends and patterns that reflect the signal's non-stationary behavior.
- Host: GitHub
- URL: https://github.com/who-else-but-arjun/tvar_elcentro
- Owner: who-else-but-arjun
- Created: 2024-12-11T14:12:47.000Z (6 months ago)
- Default Branch: main
- Last Pushed: 2024-12-11T14:25:19.000Z (6 months ago)
- Last Synced: 2025-02-14T14:57:40.822Z (3 months ago)
- Language: Jupyter Notebook
- Homepage:
- Size: 1.24 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Seismic Signal Analysis with AR Coefficients
This project analyzes seismic signals in three directions (NS, EW, and UP) using Autoregressive (AR) coefficients. The signals are processed to identify key seismic events such as primary seismic waves, secondary waves, and aftershocks.
## Dataset
The dataset used in this project is sourced from the [El Centro Seismic Station](https://www.vibrationdata.com/elcentro.htm), containing seismic signal data in the North-South (NS), East-West (EW), and Upward (UP) components.
## Project Structure
1. **Solving for 3rd order AR coefficinets**: Implementation of Yule-Walker method of solving autocorrelation matrix for AR coefficients, Linear fitting and AutoReg function from Statsmodel library.
2. **Seismic Signal Plots**: Visualization of the seismic signals over time for each component (NS, EW, and UP).
3. **AR Coefficients Cluster Maps**: Cluster maps representing the AR coefficients $\phi_1$, $\phi_2$, $\phi_3$ across time windows to understand seismic activity.
## Requirements
- Python 3.x
- Libraries: `numpy`, `matplotlib`, `scipy`, `pandas`
