# Statistical tools for teaching at NBI
This package extends some of the existing tools in
[_NumPy_](https://numpy.org) and [_SciPy_](https://scipy.org) with
some useful features designed to make life easier for the students at
the Niels Bohr Institute.
## Topics
- Reporting scientific results, including proper rounding
- Tabulation of data useful in Jupyter Notebooks
- Visualisation of data in 1 and many dimensions
- Robust calculations of sample means, variances, and covariances, for
unweighted and weighted samples. For weighted samples, both
_frequency_ and _non_-frequency weights are supported.
- Histogramming
- Sampling of arbitrary PDFs
- Curve fitting using
- Linear least squares
- Non-linear least squares
- Maximum likelihood estimates
- Extended
- Binned
- Representation of fit confidence contours
- Hyppthesis testing
- Confidence intervals
- Template fitting
- Simultaneous fitting over regions (channels)
- Likelihood calculations
## Examples of use
[This
notebook](https://cholmcc.gitlab.io/nbi-python/statistics/#nbi_stat_exa)
gives examples of use.
## Book on Statistics with Python
The book [Statistics Overview - With
Python](https://cholmcc.gitlab.io/nbi-python/statistics/#Statistik)
lays out much of the theoretical foundation for the tools available.
Some other notes on statistics is available from the same site, including
- [Principle Component Analysis](https://cholmcc.gitlab.io/nbi-python/statistics/#PCA) as a more robust alternative to boosted decision trees
- [Bootstrap and Jackknife](https://cholmcc.gitlab.io/nbi-python/statistics/#Boostrap) and why you should be careful with these estimates
- [Coefficent of determination](https://cholmcc.gitlab.io/nbi-python/statistics/#R2) and why you shouldn't use it
## Application Programming Interface Documentation
The API is
[documented](https://cholmcc.gitlab.io/nbi-python/statistics/nbi_stat).
2019 © _Christian Holm Christensen_
Raw data
{
"_id": null,
"home_page": "https://cholmcc.gitlab.io/nbi-python/statistics/#Statistik",
"name": "nbi-stat",
"maintainer": "",
"docs_url": null,
"requires_python": ">=3.6",
"maintainer_email": "",
"keywords": "Niels Bohr Institute,Physics,Statistics,Fitting,Rounding,Histogram,Random distribution,Visualizing,Tabulating",
"author": "Christian Holm Christensen",
"author_email": "cholmcc@gmail.com",
"download_url": "https://files.pythonhosted.org/packages/a2/1d/c04a2401931e1c0d410383cd02577e3993d3c90e160cdbb6640ff2fe4a22/nbi_stat-0.8.5.tar.gz",
"platform": null,
"description": "# Statistical tools for teaching at NBI \n\nThis package extends some of the existing tools in\n[_NumPy_](https://numpy.org) and [_SciPy_](https://scipy.org) with\nsome useful features designed to make life easier for the students at\nthe Niels Bohr Institute. \n\n## Topics\n\n- Reporting scientific results, including proper rounding \n- Tabulation of data useful in Jupyter Notebooks \n- Visualisation of data in 1 and many dimensions \n- Robust calculations of sample means, variances, and covariances, for\n unweighted and weighted samples. For weighted samples, both\n _frequency_ and _non_-frequency weights are supported. \n- Histogramming \n- Sampling of arbitrary PDFs \n- Curve fitting using \n - Linear least squares \n - Non-linear least squares \n - Maximum likelihood estimates \n - Extended \n - Binned \n- Representation of fit confidence contours \n- Hyppthesis testing \n- Confidence intervals \n- Template fitting \n- Simultaneous fitting over regions (channels)\n- Likelihood calculations \n\n## Examples of use \n\n[This\nnotebook](https://cholmcc.gitlab.io/nbi-python/statistics/#nbi_stat_exa)\ngives examples of use. \n\n## Book on Statistics with Python \n\nThe book [Statistics Overview - With\nPython](https://cholmcc.gitlab.io/nbi-python/statistics/#Statistik)\nlays out much of the theoretical foundation for the tools available. \n\nSome other notes on statistics is available from the same site, including \n\n- [Principle Component Analysis](https://cholmcc.gitlab.io/nbi-python/statistics/#PCA) as a more robust alternative to boosted decision trees \n- [Bootstrap and Jackknife](https://cholmcc.gitlab.io/nbi-python/statistics/#Boostrap) and why you should be careful with these estimates \n- [Coefficent of determination](https://cholmcc.gitlab.io/nbi-python/statistics/#R2) and why you shouldn't use it \n\n## Application Programming Interface Documentation \n\nThe API is\n[documented](https://cholmcc.gitlab.io/nbi-python/statistics/nbi_stat). \n\n2019 \u00a9 _Christian Holm Christensen_\n",
"bugtrack_url": null,
"license": "GPL",
"summary": "Statistics tools for teaching at NBI",
"version": "0.8.5",
"project_urls": {
"Documentation": "https://cholmcc.gitlab.io/nbi-python/statistics/nbi_stat",
"Homepage": "https://cholmcc.gitlab.io/nbi-python/statistics/#Statistik",
"Source Code": "https://gitlab.com/cholmcc/nbi-python"
},
"split_keywords": [
"niels bohr institute",
"physics",
"statistics",
"fitting",
"rounding",
"histogram",
"random distribution",
"visualizing",
"tabulating"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "ea936c2442b2e322c4313392318501082bc83276f1c4c65127a34f1961c1bd88",
"md5": "5829ae147436752dc6016c07c887984f",
"sha256": "db2cc99e658430180a28c952414b591d115e3724fa29578d84fd4793d62d1312"
},
"downloads": -1,
"filename": "nbi_stat-0.8.5-py3-none-any.whl",
"has_sig": false,
"md5_digest": "5829ae147436752dc6016c07c887984f",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.6",
"size": 56051,
"upload_time": "2024-02-22T18:21:54",
"upload_time_iso_8601": "2024-02-22T18:21:54.159812Z",
"url": "https://files.pythonhosted.org/packages/ea/93/6c2442b2e322c4313392318501082bc83276f1c4c65127a34f1961c1bd88/nbi_stat-0.8.5-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "a21dc04a2401931e1c0d410383cd02577e3993d3c90e160cdbb6640ff2fe4a22",
"md5": "28175b4fa56bb414d575cbdea321a1dc",
"sha256": "98d055c9b9f2b4eb8bd14ce6dce669d511e5647f51c7d0bd855401ab4e5f93f9"
},
"downloads": -1,
"filename": "nbi_stat-0.8.5.tar.gz",
"has_sig": false,
"md5_digest": "28175b4fa56bb414d575cbdea321a1dc",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.6",
"size": 56943,
"upload_time": "2024-02-22T18:21:56",
"upload_time_iso_8601": "2024-02-22T18:21:56.422349Z",
"url": "https://files.pythonhosted.org/packages/a2/1d/c04a2401931e1c0d410383cd02577e3993d3c90e160cdbb6640ff2fe4a22/nbi_stat-0.8.5.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-02-22 18:21:56",
"github": false,
"gitlab": true,
"bitbucket": false,
"codeberg": false,
"gitlab_user": "cholmcc",
"gitlab_project": "nbi-python",
"lcname": "nbi-stat"
}
JSON