---
myst:
    substitutions:
        bloc1:
            python: 1-7,17-104
            cpp: 3-8,13-87
        bloc2:
            python: 39-42
            cpp: 22-25
---
(use_case_python_macro_dataserver_compute_quantile)=
# Macro "**{{"dataserverComputeQuantile." + extension}}**"

## Objective

The objective of this macro is to test the classical quantile estimation and compare it to the Wilks 
estimation for a dummy gaussian distribution. Four different estimations of the 95% quantile value 
are done (along with one estimation of th 99% quantile) for illustration purposes:

- using the usual method with a 200-points sample.
- using the usual method with a 400-points sample.
- using the Wilks method with a 95% confidence level (with 59-points sample).
- using the Wilks method with a 95% confidence level (with 400-points sample).
- using the Wilks method with a 99% confidence level (with 90-points sample).

## Macro {{uranie}}

{{
    "```{literalinclude} " + parent_dir
    + "/roottest/uranie/doc/dataserver/use_cases/" + language + "/dataserverComputeQuantile." 
    + extension 
    + "\n"
    + ":language: " + language + "\n"
    + ":lines: " + bloc1[language] + "\n"
    + "```"
}}

In this macro, a dummy dataserver is created with a single attribute named "x". Four histograms are 
prepared to store the resulting value. Then the same loop will be used to computed 10000 values of 
every quantile with different switchs to use one method instead of the other, or to change the number 
of points in the sample and/or the confidence level. All this is defined in the small part before 
the loop:

{{
    "```{literalinclude} " + parent_dir
    + "/roottest/uranie/doc/dataserver/use_cases/" + language + "/dataserverComputeQuantile." 
    + extension 
    + "\n"
    + ":language: " + language + "\n"
    + ":lines: " + bloc2[language] + "\n"
    + "```"
}}

Then the computation is performed, first for the usual method (first two iterations of `iq`), then 
for the Wilks estimation (last two iterations of `iq`). Every computational result is stored in the 
corresponding histogram which is finally displayed and shown in the following subsection.

## Graph

{{
    "```{" "figure" "} " + parent_dir +
        "/roottest/build/uranie/doc/dataserver/use_cases/" + language + "/dataserverComputeQuantile.png\n"
    ":align: center\n"
    ":name: usecases_dataserverComputeQuantile\n"
    + figure_scale_reduced + "\n"
    "\n"
    "Graph of the macro `\"dataserverComputeQuantile." + extension + "\"`\n"
    "```"
}}