diff --git a/source/statistics/figures/normal_law_tails.svg b/source/statistics/figures/normal_law_tails.svg new file mode 100644 index 0000000..13f05c4 --- /dev/null +++ b/source/statistics/figures/normal_law_tails.svg @@ -0,0 +1,416 @@ + + + +image/svg+xml LaDistribucióNormalLaDistribuciónNormalValorsValores ProbabilitatProbabilidad 95% de valors95% de valores 99% de valors99% de valores Desviació Estàndardde la MitjanaDesviación Estándarde la Media Valors ZValores Z Probabilitat de Casosen segments de la corbaProbabilidad de Casosen porciones de la curva % Acumulatiu% Acumulativo Valors TValores T Left tailRight tail diff --git a/source/statistics/tests/parametric/ztest.rst b/source/statistics/tests/parametric/ztest.rst index 82c3586..20dc042 100644 --- a/source/statistics/tests/parametric/ztest.rst +++ b/source/statistics/tests/parametric/ztest.rst @@ -2,7 +2,7 @@ Z-Test ------- The z-test is used to assess if the mean :math:`\overline{x}` of sample :math:`X` significantly differ from the one of a known population. -The *significance level* is determined by a *p-value* threshold. +The *significance level* is determined by a *p-value* threshold chosen prior doing the test. Conditions for using a z-test: @@ -14,12 +14,10 @@ Conditions for using a z-test: According to central limit theorem, a distribution is well approximated when reaching 30 samples. See `here `__ for more infos. -One-tailed vs Two-tailed -======================== To perform a z-test, you should compute the *standard score* (or *z-score*) of your sample. -It corresponds to the projection of the sample mean :math:`\overline{x}` under the original population distribution. +It characterizes how far from the population mean :math:`\mu` your sample mean :math:`\overline{x}` is, in unit of standard deviation :math:`\sigma`. It is computed as follow: .. math:: @@ -31,10 +29,34 @@ It is computed as follow: .. math:: Z=\frac{\overline{x}-\mu}{\mathrm{SEM}}=\frac{\overline{x}-\mu}{\frac{s}{\sqrt{n}}} - This formula originate from the t-test and :math:`Z` technically follow a t-distribution. - However, if :math:`n` is sufficiently large, the sample distribution is very close to a normal one. - So close that, using the normal in place of the student-t to compute p values leads to nominal differences (`source `__). - + In this case, :math:`Z` technically follow a t-distribution (student test). + However, if :math:`n` is sufficiently large, the distribution followed by :math:`Z` is very close to a normal one. + So close that, using z-test in place of the student test to compute *p-values* leads to nominal differences (`source `__). + +From :math:`Z`, the z-test *p-value* can be derived using the :math:`\mathcal{N}(0,1)` :ref:`CDF `. +That *p-value* is computed as follow: + +* Left "tail" of the :math:`\mathcal{N}(0,1)` distribution: + + .. math:: + \alpha=P(\mathcal{N}(0,1)