science-notes/source/statistics/notations.rst
2023-10-17 19:44:12 +02:00

72 lines
2 KiB
ReStructuredText

Notations
---------------
This page was inspired from the following `Wikipedia page <https://en.wikipedia.org/wiki/Notation_in_probability_and_statistics>`_.
Sample and Population
=====================
A *population* is the entire population that you want to analyze. It is a exhaustive set. A *sample* is a subset of the population. It is a non-exhaustive set.
.. figure:: figures/population_sample.svg
:align: center
:width: 300px
*Inferential statistics* allows you draw conclusions (with a certain degree of confidence) on a population using samples.
A Tale of Schizophrenia
================================
Two different notation conventions are used. The one to use depends if you are working on a *population* or a *sample*:
.. list-table:: Notation
:align: center
:header-rows: 1
* - Metric
- Population
- Sample
- Notes
* - Sample mean
- :math:`\mu`
- :math:`\overline{x}`
-
* - Variance
- :math:`\sigma^2`
- :math:`s^2`
- :math:`s^2_n` without `Bessel's Correction <bessel_correction>`__
* - Standard deviation
- :math:`\sigma`
- :math:`s`
- :math:`s_n` without `Bessel's Correction <bessel_correction>`__
To determine the metric of a population (say :math:`\mu`) using a sample, we use an estimator.
In the case of :math:`\mu`, we use :math:`\overline{x}` as an estimator.
.. note::
Estimators can also be denoted with the hat symbol.
For example :math:`\hat{\mu}\equiv\overline{x}`.
Operators
==========
Expected value
^^^^^^^^^^^^^^^
The expected value (*espérance*) of a random variable :math:`X` is noted :math:`\mathbb{E}[X]`.
It as the following linearity properties:
.. math::
\mathbb{E}[X+Y]=\mathbb{E}[X] + \mathbb{E}[Y]
.. math::
\mathbb{E}[\alpha X]=\alpha\mathbb{E}[X]
Variance
^^^^^^^^
The variance operator of a random variable :math:`X` is noted :math:`\mathbb{V}[X]` or :math:`\mathrm{Var}[X]`.
Coraviance
^^^^^^^^^^
The covariance operator of a random variable :math:`X` and :math:`Y`
is noted :math:`\mathrm{Cov}[X,Y]`.