Changed info file output, added setup, conftest, auxiliary test modul…

…e, changed unit tests ...

docs/credits.rst

@@ -14,9 +14,10 @@ Development Lead
 Contributors
 ^^^^^^^^^^^^
 
-`Sebastian Becker <https://github.com/sebecker>`_
+`Tobias Raabe <https://github.com/tobiasraabe>`_, `Sebastian Becker <https://github.com/sebecker>`_
 
 Acknowledgments
 ^^^^^^^^^^^^^^^
 
-We appreciate the financial support of the `AXA Research Fund <https://www.axa-research.org/>`_ and the  `University of Bonn <https://www.uni-bonn.de>`_. We are indebted to the open source community as we build on top of numerous open source tools such as the `SciPy Stack <https://www.scipy.org>`_ and `statsmodels <http://statsmodels.sourceforge.net/>`.
+We appreciate the financial support of the `AXA Research Fund <https://www.axa-research.org/>`_ and the  `University of Bonn <https://www.uni-bonn.de>`_. We are indebted to the open source community as we build on top of numerous open source tools such as the `SciPy Stack <https://www.scipy.org>`_ and
+`statsmodels <http://statsmodels.sourceforge.net/>`_.

docs/tutorial.rst

@@ -11,4 +11,70 @@ Parametric Assumptions
 Model Specification
 -------------------
 
-The model is specified in an initialization file.
+The model is specified in an initialization file. The specific file includes the following sections:
+
+
+**SIMULATION**
+
+The *SIMULATION* part contains basic configurations that determine the simulation and output process.
+
+=======     ======      ==================
+Key         Value       Interpretation
+=======     ======      ==================
+agents      int         number of individuals
+seed        int         seed for the specific simulation
+source      str         name for output files
+=======     ======      ==================
+
+**TREATED & UNTREATED**
+
+The *TREATED* and *UNTREATED* paragraph are similar regarding their structure. Both contain the parameters that determine the expected wage dependent on the treatment status. There can be added as many covariates as wished, but the number has to be the same for both cases.
+
+*Note:* The first coefficient is interpreted as an intercept.
+
+=======     ======      ==================
+Key         Value       Interpretation
+=======     ======      ==================
+coeff       float       intercept coefficient
+coeff       float       coefficient of the first covariate
+coeff       float       coefficient of the second covariate
+ ...
+=======     ======      ==================
+
+**COST**
+
+*COST* includes parameters related to the cost function.
+
+*Note:* The first coefficient is interpreted as an intercept.
+
+=======     ======      ==================
+Key         Value       Interpretation
+=======     ======      ==================
+coeff       float       intercept coefficient
+coeff       float       coefficient of the first covariate
+coeff       float       coefficient of the second covariate
+ ...
+=======     ======      ==================
+
+**DIST**
+
+The *DIST* Section determines distributional characteristics regarding the unobservables.
+
+*Note:* The coefficients
+
+=======     ======      ==================
+Key         Value       Interpretation
+=======     ======      ==================
+coeff       float      :math:`\sigma_{1}`
+coeff       float      :math:`\sigma_{12}`
+coeff       float      :math:`\sigma_{13}`
+coeff       float      :math:`\sigma_{2}`
+coeff       float      :math:`\sigma_{23}`
+coeff       float      :math:`\sigma_{3}`
+=======     ======      ==================
+
+
+
+
+
+

grmpy/__init__.py

@@ -0,0 +1,24 @@
+"""The module allows to run tests from inside the interpreter
+"""
+import os
+
+try:
+    import pytest
+except ImportError:
+    pass
+
+__version__ = '1.0.0'
+
+
+def test():
+    """ Run PYTEST for the package.
+    """
+
+    package_directory = os.path.dirname(os.path.realpath(__file__))
+    current_directory = os.getcwd()
+
+    os.chdir(package_directory)
+
+    pytest.main()
+
+    os.chdir(current_directory)

grmpy/read/read.py

@@ -1,4 +1,5 @@
-
+"""The module contains the main function of the init file import process
+"""
 
 import shlex
 
@@ -7,7 +8,8 @@
 
 
 def read(file_):
-    """reads the initialization file and provides an dictionary with parameters for the simulation"""
+    """The function reads the initialization file and returns a dictionary with parameters for the simulation
+    """
     dict_ = {}
     for line in open(file_).readlines():
 

grmpy/read/read_auxiliary.py

@@ -1,8 +1,11 @@
+"""This module provides auxiliary functions for the import process of the init file.
+"""
+
 import numpy as np
 
 
 def process(list_, dict_, keyword):
-    """processes keyword parameters"""
+    """The function processes keyword parameters and creates dictionary elements."""
     name, val = list_[0], list_[1]
 
     if name not in dict_[keyword].keys() and name in ['coeff']:
@@ -25,7 +28,8 @@ def process(list_, dict_, keyword):
 
 
 def auxiliary(dict_):
-    """
+    """The function creates an new dictionary entry 'AUX' that includes starting values of
+    each parameter and the number of covariates.
     """
     dict_['AUX'] = {}
     if dict_['DIST']['coeff'] == [0.0] * len(dict_['DIST']['coeff']):
@@ -41,8 +45,6 @@ def auxiliary(dict_):
             dict_[key_]['all'] = dict_[key_]['coeff']
             dict_[key_]['all'] = np.array(dict_[key_]['all'])
 
-    # Create keys that contain all standard deviation and covariance parameters
-
     # Number of covariates
     num_covars_out = len(dict_['TREATED']['all'])
     num_covars_cost = len(dict_['COST']['all'])

grmpy/setup.py

@@ -0,0 +1,11 @@
+from setuptools import setup
+
+setup(name='grmpy',
+      version='0.1',
+      description='grmpy is a package for simulation and analysis of economic aspects of the generalized roy model',
+      url='https://github.com/grmToolbox/grmpy',
+      author='Phillip Eisenhauer',
+      author_email='eisenhauer@policy-lab.org',
+      license='MIT',
+      packages=['numpy', 'scipy', 'pytest', 'pandas', 'statsmodels'],
+      zip_safe=False)

grmpy/simulation/simulation.py → grmpy/simulate/simulate.py

@@ -1,9 +1,12 @@
+"""The module provides the simulation process.
+"""
 import numpy as np
+import os.path
 
-from grmpy.simulation.simulation_auxiliary import simulate_unobservables
-from grmpy.simulation.simulation_auxiliary import simulate_outcomes
-from grmpy.simulation.simulation_auxiliary import write_output
-from grmpy.simulation.simulation_auxiliary import print_info
+from grmpy.simulate.simulate_auxiliary import simulate_unobservables
+from grmpy.simulate.simulate_auxiliary import simulate_outcomes
+from grmpy.simulate.simulate_auxiliary import write_output
+from grmpy.simulate.simulate_auxiliary import print_info
 from grmpy.read.read import read
 
 
@@ -12,8 +15,10 @@ def simulate(init_file):
     """
 
     # Transform init file to dictionary
-
+    assert isinstance(init_file, str)
+    assert os.path.isfile(init_file)
     init_dict = read(init_file)
+
     # Distribute information
     is_deterministic = init_dict['DETERMINISTIC']
     num_agents = init_dict['SIMULATION']['agents']
@@ -44,17 +49,15 @@ def simulate(init_file):
     Z[:, 0], X[:, 0] = 1.0, 1.0
 
     # Simulate unobservables
-    # Read information about the distribution and the specific means from the init dic
-
     U, V = simulate_unobservables(covar_, vars_, num_agents)
 
     # Simulate endogeneous variables
-
     Y, D, Y_1, Y_0 = simulate_outcomes([X, Z], U, coeffs)
 
     # Write output file
-    df = write_output([Y, D, Y_1, Y_0], [X, Z], [U, V], source, is_deterministic)
+    df = write_output([Y, D, Y_1, Y_0], [X, Z], [U, V],
+                      source, is_deterministic)
 
     print_info(df, [Y0_coeffs, Y1_coeffs, C_coeffs, Dist_coeffs], source)
 
-    return df, Y, Y_1, Y_0, D, X, Z, U, V
+    return df