https://github.com/stephanielees/ingarch-countts
INGARCH process for count time series
https://github.com/stephanielees/ingarch-countts
Last synced: 3 months ago
JSON representation
INGARCH process for count time series
- Host: GitHub
- URL: https://github.com/stephanielees/ingarch-countts
- Owner: stephanielees
- Created: 2024-12-28T08:49:27.000Z (5 months ago)
- Default Branch: main
- Last Pushed: 2025-01-09T05:54:40.000Z (5 months ago)
- Last Synced: 2025-01-09T06:38:18.373Z (5 months ago)
- Language: Jupyter Notebook
- Size: 2.64 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# INGARCH-countTS
INGARCH process for count time series
The demo video is [here](https://youtu.be/89e2Nley1uo)
Discrete time series can be found in many applications, and the nature of such time series is different from the continuous time series. Therefore, we need a different way to find the best process that could generate count time series. Integer-valued GARCH (INGARCH) is one of various models for modelling (nonnegative) discrete time series.
What's in this repository?
1. `datasource-wikipedia-math.ipynb`: This is the notebook describing how I got the original data, which is time series of graphs (more technically, it's called a spatiotemporal type of data), and extract the data from one node.
2. `wiki_BucketSort.csv`: The output of the file above. This csv file contains the time series of Bucket Sort in Wikipedia and of the pages which mention Bucket sort (the covariates).
3. `CountTS-INGARCH.Rmd`: The main file.
I initially wanted to use Python for all the steps in this project. In fact, I tried using another model that I studied first before INGARCH. But I got stuck in the coding stage 😅, so I changed my plan (for now) to using INGARCH, which has many implementation packages written in R.