# clumpi [klˈʌmpάɪ]
## Overview
A simple python package to calculate Clumpiness for RMFC analysis by Zhang, Bradlow & Small (2015).
Easy use with `clumpi.get_RFC()`
## Requirements
- Python
- pandas
- numpy
works well with Google Colab.
## Installation
```bash
pip install git+https://github.com/jniimi/clumpi.git
```
## Dataset
Use your time-series event data with ID and time.
- Create DataFrame that records only the point in time when the event occurred in the time series data.
- The name of the variables can be anything.
| user_id | t |
|:--------|--------:|
| Ava | 1 |
| Ava | 4 |
| ... | ... |
| Jack | 3 |
| Jack | 10 |
| ... | ... |
Check out our sample dataset for further details.
```python
df = clumpi.load_sample_data()
display(df)
```
## Usage
### Log to Clumpiness
Use the function `clumpi.get_RFC()` to calculate. Specify following information for the arguments.
- `id`: a var name in df indicating user
- `t`: a var name in df indicating time
- `N`: total number of events can occur during the period
- `M` (optional): a number of iterations for the simulation to calculate threshold (3000 for default)
- `alpha` (optional): significance probability for the test of regularity (0.05 for default)
### Simply Calculate H0
Use the function `clumpi.calc_threshold()` to calculate upper `alpha` % point in `M` times simulation.
All you need to specify are `N`, `M`, and `alpha` (See `clumpi.get_RFC`).
# Acknoledgement
The simulation in this package is based on Appendix B by Zhang et al. (2015).
Zhang, Y., Bradlow, E. T., & Small, D. S. (2015). Predicting customer value using clumpiness: From RFM to RFMC. Marketing Science, 34(2), 195-208.
https://doi.org/10.1287/mksc.2014.0873
# Author
jniimi ([@JvckAndersen](https://twitter.com/JvckAndersen))
Raw data
{
"_id": null,
"home_page": "https://github.com/jniimi/clumpi",
"name": "clumpi",
"maintainer": "",
"docs_url": null,
"requires_python": "",
"maintainer_email": "",
"keywords": "RFMC Clumpiness marketing CRM",
"author": "jniimi",
"author_email": "jniimi@meijo-u.ac.jp",
"download_url": "https://files.pythonhosted.org/packages/c2/38/88f37b26f3415baaa6594ca74b03342c970cfabe4a622acb17d484d7d208/clumpi-0.0.1.tar.gz",
"platform": null,
"description": "# clumpi [kl\u02c8\u028cmp\u03ac\u026a]\n\n## Overview\nA simple python package to calculate Clumpiness for RMFC analysis by Zhang, Bradlow & Small (2015).\nEasy use with `clumpi.get_RFC()`\n\n## Requirements\n- Python\n- pandas\n- numpy\n\nworks well with Google Colab.\n\n## Installation\n```bash\npip install git+https://github.com/jniimi/clumpi.git\n```\n## Dataset\nUse your time-series event data with ID and time. \n- Create DataFrame that records only the point in time when the event occurred in the time series data. \n- The name of the variables can be anything.\n\n| user_id | t |\n|:--------|--------:|\n| Ava | 1 |\n| Ava | 4 |\n| ... | ... |\n| Jack | 3 |\n| Jack | 10 |\n| ... | ... |\n\nCheck out our sample dataset for further details.\n```python\ndf = clumpi.load_sample_data()\ndisplay(df)\n```\n\n\n## Usage\n### Log to Clumpiness\nUse the function `clumpi.get_RFC()` to calculate. Specify following information for the arguments.\n- `id`: a var name in df indicating user\n- `t`: a var name in df indicating time\n- `N`: total number of events can occur during the period\n- `M` (optional): a number of iterations for the simulation to calculate threshold (3000 for default)\n- `alpha` (optional): significance probability for the test of regularity (0.05 for default)\n\n### Simply Calculate H0\nUse the function `clumpi.calc_threshold()` to calculate upper `alpha` % point in `M` times simulation. \n\nAll you need to specify are `N`, `M`, and `alpha` (See `clumpi.get_RFC`).\n\n# Acknoledgement\nThe simulation in this package is based on Appendix B by Zhang et al. (2015).\n\nZhang, Y., Bradlow, E. T., & Small, D. S. (2015). Predicting customer value using clumpiness: From RFM to RFMC. Marketing Science, 34(2), 195-208.\nhttps://doi.org/10.1287/mksc.2014.0873\n\n# Author\njniimi ([@JvckAndersen](https://twitter.com/JvckAndersen))\n",
"bugtrack_url": null,
"license": "MIT",
"summary": "Calculate Clumpiness index by Zhang, Bradlow and Small (2015)",
"version": "0.0.1",
"project_urls": {
"Download": "https://github.com/jniimi/clumpi",
"Homepage": "https://github.com/jniimi/clumpi"
},
"split_keywords": [
"rfmc",
"clumpiness",
"marketing",
"crm"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "c23888f37b26f3415baaa6594ca74b03342c970cfabe4a622acb17d484d7d208",
"md5": "2070ed20aed9cdea9a8aadff04759983",
"sha256": "31d2745f185797551581931fb1bb0cee7ad01be18ade7ee845ba19b8e2f0e94d"
},
"downloads": -1,
"filename": "clumpi-0.0.1.tar.gz",
"has_sig": false,
"md5_digest": "2070ed20aed9cdea9a8aadff04759983",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 4229,
"upload_time": "2023-09-02T13:08:56",
"upload_time_iso_8601": "2023-09-02T13:08:56.672048Z",
"url": "https://files.pythonhosted.org/packages/c2/38/88f37b26f3415baaa6594ca74b03342c970cfabe4a622acb17d484d7d208/clumpi-0.0.1.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2023-09-02 13:08:56",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "jniimi",
"github_project": "clumpi",
"travis_ci": false,
"coveralls": false,
"github_actions": false,
"requirements": [],
"lcname": "clumpi"
}
JSON
