----------------------------------------------
lines 5-81 of file: example/user/hold_out_1.py
----------------------------------------------

# {xrst_begin user_hold_out_1.py}
# {xrst_spell
#     exp
# }
# {xrst_comment_ch #}
#
# Using hold_out in Data, Subset Data, and Option Tables
# ######################################################
#
# Purpose
# *******
# This example shows how to use hold_out in the
# :ref:`data_table<data_table@hold_out>` ,
# :ref:`option_table<option_table@hold_out_integrand>` , and
# :ref:`hold_out_command-name` .
#
# Integrands
# **********
# For this example there are two integrand,
# ``Sincidence`` and ``prevalence`` .
#
# Nodes
# *****
# The node table is set up so that there are lots of child nodes
# (with no data. This makes sure that the data being fit gets
# evenly distributed between the nodes that do have data and
# reaches :ref:`hold_out_command@max_fit` .
#
# Data
# ****
#
# prevalence
# ==========
# All of the prevalence data is zero, but it is held out using
# :ref:`option_table@hold_out_integrand` .
#
# Sincidence
# ==========
# There are many incidence data points.
# The first Sincidence data value is zero and it is held out using
# the data table :ref:`data_table@hold_out` equal to one.
# The other Sincidence data are
# the true value for incidence and have the data table hold_out
# equal to zero.
# There are four data points for each node, except the root node n0.
# The :ref:`hold_out_command-name` is used to randomly select a subset of these
# points to be held out using max_fit equal to two time the number of nodes.
#
# Model
# *****
# There is only one rate *iota* and it is constant in age and time.
# The corresponding model for the Sincidence data is *iota* .
# The corresponding mode for the prevalence data is
# 1 ``- exp`` ( *iota* * *age* ) .
#
# Fit
# ***
# Because the zero prevalence data and zero incidence data is held out,
# the fitting value for *iota* is very close to the true value
# for incidence.
#
# Parameters
# **********
# you can change the following parameters in this test:
# {xrst_literal
#     BEGIN_PARAMETERS
#     END_PARAMETERS
# }
#
# Source Code
# ***********
# {xrst_literal
#     BEGIN PYTHON
#     END PYTHON
# }
#
# {xrst_end user_hold_out_1.py}