Reliable Statistics
===================
Statistics often get bad rap for being inaccurate or misleading. They are after all
guesses. We can improve their quality by adding information about confidence in
these numbers. This project provides tools to compute the confidence levels.
Example of usage in real-life situations is reliability engineering.
Reliability engineering deals with estimating parameters or qualities of a product or
process or experiment. For simplicity, we assume that all units of a product or results
of an experiment are random variables. Collectively, let's call them samples.
We assume that the samples are independent (one sample has no effect on another
sample) and identically distributed (the reliability or properties of underlying
random variable stay the same for each sample).
Concepts
--------
* Reliability is probability of success. The math assumes infinite number of samples,
but we can get access to only a finite number of samples. Therefore, we can compute
only an estimate of the actual reliability. Based on the number of samples, we
qualify the quality of this estimate using *confidence*.
* Confidence in reliability is probability that the actual reliability of the
population is at least the provided reliability level.
For example, we can say "If we see zero failures in 10 samples of a success-failure
experiment, we have 95% confidence that the reliability is at least about 74%".
* Assurance simplifies reliability and confidence by setting both of them the same.
The result is just one number that is easier to communicate. For example, 90%
assurance means 90% reliability with 90% confidence. Given the number of samples
and number of failures, assurance is just one number.
This library provides methods to calculate these statistics for infinite and finite
population sizes.
Example usage in a python file:
.. code-block:: python
from relistats.binomial import assurance
n = 22
a = assurance(n, 0) or 0
print(f"Assurance at {n} good samples: {a*100:.1f}%")
References
----------
* S.M. Joshi, "Computation of Reliability Statistics for Success-Failure Experiments,"
`arXiv:2303.03167 [stat.ME] <https://doi.org/10.48550/arXiv.2303.03167>`_, March 2023.
* S.M. Joshi, "Computation of Reliability Statistics for Finite Samples of Success-Failure Experiments,"
`arXiv:2305.16578 [cs.IT] <http://arxiv.org/abs/2305.16578>`_, May 2023.
* Jupyter notebooks showing how to use this library
`relistats_notebook <https://github.com/sanjaymjoshi/relistats_notebook>`_
* Interactive online versions of the Jupyter notebook on Google Colab.
* `Infinite population <https://colab.research.google.com/github/sanjaymjoshi/relistats_notebook/blob/main/relistats_binomial.ipynb>`_.
* `Finite population <https://colab.research.google.com/github/sanjaymjoshi/relistats_notebook/blob/main/relistats_binomial_finite.ipynb>`_.
Additional documentation.
* `Usage <docs/source/usage.rst>`_ for installation and how to use.
* `Background <docs/source/background.rst>`_ for concepts and mathematical background.
* `CHANGELOG.md` for revision history.
Credits
----------
This package was originally created with Cookiecutter and the
`sourcery-ai/python-best-practices-cookiecutter
<https://github.com/sourcery-ai/python-best-practices-cookiecutter>`_
project template. Later modified by author.
Raw data
{
"_id": null,
"home_page": "",
"name": "relistats",
"maintainer": "",
"docs_url": null,
"requires_python": ">=3.10",
"maintainer_email": "Sanjay Joshi <sanjaymjoshi@iitbombay.org>",
"keywords": "assurance,confidence,percentile,reliability,statistics",
"author": "",
"author_email": "Sanjay Joshi <sanjaymjoshi@iitbombay.org>",
"download_url": "https://files.pythonhosted.org/packages/89/6d/b611ef37df5dc0937ea55aa20d6d686acad9ce2ddbb5a8040db16ce21bbd/relistats-1.2.0.tar.gz",
"platform": null,
"description": "Reliable Statistics\n===================\n\nStatistics often get bad rap for being inaccurate or misleading. They are after all\nguesses. We can improve their quality by adding information about confidence in\nthese numbers. This project provides tools to compute the confidence levels.\nExample of usage in real-life situations is reliability engineering.\n\nReliability engineering deals with estimating parameters or qualities of a product or\nprocess or experiment. For simplicity, we assume that all units of a product or results\nof an experiment are random variables. Collectively, let's call them samples.\nWe assume that the samples are independent (one sample has no effect on another\nsample) and identically distributed (the reliability or properties of underlying\nrandom variable stay the same for each sample).\n\nConcepts\n--------\n* Reliability is probability of success. The math assumes infinite number of samples,\n but we can get access to only a finite number of samples. Therefore, we can compute\n only an estimate of the actual reliability. Based on the number of samples, we\n qualify the quality of this estimate using *confidence*.\n\n* Confidence in reliability is probability that the actual reliability of the\n population is at least the provided reliability level. \n For example, we can say \"If we see zero failures in 10 samples of a success-failure\n experiment, we have 95% confidence that the reliability is at least about 74%\".\n\n* Assurance simplifies reliability and confidence by setting both of them the same.\n The result is just one number that is easier to communicate. For example, 90%\n assurance means 90% reliability with 90% confidence. Given the number of samples\n and number of failures, assurance is just one number.\n\nThis library provides methods to calculate these statistics for infinite and finite\npopulation sizes.\n\nExample usage in a python file:\n\n.. code-block:: python\n\n from relistats.binomial import assurance\n \n n = 22\n a = assurance(n, 0) or 0\n print(f\"Assurance at {n} good samples: {a*100:.1f}%\")\n\nReferences\n----------\n* S.M. Joshi, \"Computation of Reliability Statistics for Success-Failure Experiments,\"\n `arXiv:2303.03167 [stat.ME] <https://doi.org/10.48550/arXiv.2303.03167>`_, March 2023.\n* S.M. Joshi, \"Computation of Reliability Statistics for Finite Samples of Success-Failure Experiments,\"\n `arXiv:2305.16578 [cs.IT] <http://arxiv.org/abs/2305.16578>`_, May 2023.\n* Jupyter notebooks showing how to use this library\n `relistats_notebook <https://github.com/sanjaymjoshi/relistats_notebook>`_\n* Interactive online versions of the Jupyter notebook on Google Colab.\n\n * `Infinite population <https://colab.research.google.com/github/sanjaymjoshi/relistats_notebook/blob/main/relistats_binomial.ipynb>`_.\n * `Finite population <https://colab.research.google.com/github/sanjaymjoshi/relistats_notebook/blob/main/relistats_binomial_finite.ipynb>`_.\n\nAdditional documentation.\n\n* `Usage <docs/source/usage.rst>`_ for installation and how to use.\n* `Background <docs/source/background.rst>`_ for concepts and mathematical background.\n* `CHANGELOG.md` for revision history.\n\nCredits\n----------\nThis package was originally created with Cookiecutter and the\n`sourcery-ai/python-best-practices-cookiecutter\n<https://github.com/sourcery-ai/python-best-practices-cookiecutter>`_\nproject template. Later modified by author.\n",
"bugtrack_url": null,
"license": "MIT License",
"summary": "Reliable Statistics",
"version": "1.2.0",
"project_urls": {
"Bug Tracker": "https://github.com/sanjaymjoshi/relistats/issues",
"Changelog": "https://github.com/sanjaymjoshi/relistats/blob/main/CHANGELOG.md",
"Documentation": "https://relistats.readthedocs.io/en/latest/",
"Homepage": "https://github.com/sanjaymjoshi/relistats",
"Repository": "https://github.com/sanjaymjoshi/relistats"
},
"split_keywords": [
"assurance",
"confidence",
"percentile",
"reliability",
"statistics"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "4cd6a796d2885f2d633a68859cca96f509036ece8d952cdce467888fb95fdf4a",
"md5": "ea043ece17b54a321097ff6fd252c79c",
"sha256": "9dc56e568be667055867f158048c4e5d69931f10c711c44fce806468ef01431f"
},
"downloads": -1,
"filename": "relistats-1.2.0-py3-none-any.whl",
"has_sig": false,
"md5_digest": "ea043ece17b54a321097ff6fd252c79c",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.10",
"size": 11564,
"upload_time": "2024-03-08T10:39:44",
"upload_time_iso_8601": "2024-03-08T10:39:44.609823Z",
"url": "https://files.pythonhosted.org/packages/4c/d6/a796d2885f2d633a68859cca96f509036ece8d952cdce467888fb95fdf4a/relistats-1.2.0-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "896db611ef37df5dc0937ea55aa20d6d686acad9ce2ddbb5a8040db16ce21bbd",
"md5": "086c64fe72c9565b38e7fcaf41dc5cc2",
"sha256": "51281cb0ee69111f6b9f13a326475a0222cd9a4d5dd81a2621a7e058c8b390d9"
},
"downloads": -1,
"filename": "relistats-1.2.0.tar.gz",
"has_sig": false,
"md5_digest": "086c64fe72c9565b38e7fcaf41dc5cc2",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.10",
"size": 543520,
"upload_time": "2024-03-08T10:39:46",
"upload_time_iso_8601": "2024-03-08T10:39:46.039697Z",
"url": "https://files.pythonhosted.org/packages/89/6d/b611ef37df5dc0937ea55aa20d6d686acad9ce2ddbb5a8040db16ce21bbd/relistats-1.2.0.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-03-08 10:39:46",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "sanjaymjoshi",
"github_project": "relistats",
"travis_ci": false,
"coveralls": true,
"github_actions": true,
"lcname": "relistats"
}
JSON
