.. _book_fig_chapter5_fig_poisson_likelihood:


Binned Regression: Poisson vs Gaussian
--------------------------------------
Figure 5.15

The left panels show data sets with 50 points, binned in 5 bins (upper panels)
and 40 bins (lower panels). The curves show the input distribution (solid), the
Poisson solution (dashed), and the Gaussian solution (dotted). The right panels
show 1-sigma, 2-sigma, and 3-sigma likelihood contours for eqs. 5.91 (dark
lines) and 5.90 (light lines). With 5 bins (top row) there are enough counts
in each bin so that the Gaussian and Poisson predictions are very similar. As
the number of bins is increased, the counts decrease and the Gaussian
approximation becomes biased.



.. image:: ../images/chapter5/fig_poisson_likelihood_1.png
    :scale: 100
    :align: center


.. raw:: html
    
    <div class="toggle_trigger"><a href="#">

**Code output:**

.. raw:: html

    </a></div>
    <div class="toggle_container">

.. literalinclude:: fig_poisson_likelihood.txt

.. raw:: html

    </div>
    <div class="toggle_trigger" id="start_open"><a href="#">

**Python source code:**

.. raw:: html

    </a></div>
    <div class="toggle_container">

.. literalinclude:: fig_poisson_likelihood.py
    :lines: 15-

.. raw:: html

    </div>
    <div align="right">

:download:`[download source: fig_poisson_likelihood.py] <fig_poisson_likelihood.py>`

.. raw:: html
    
    </div>