Add pandas as hard dependency

Author: kyleam

readme.md

@@ -46,11 +46,13 @@ Another option is to clone the repository and install PyMC using `python setup.p
 
 ## Dependencies
 
-PyMC is tested on Python 2.7 and 3.3 and depends on Theano, NumPy, SciPy, and Matplotlib (see setup.py for version information).
+PyMC is tested on Python 2.7 and 3.3 and depends on Theano, NumPy,
+SciPy, Pandas, and Matplotlib (see setup.py for version information).
 
 ### Optional
 
-The GLM submodule relies on Pandas and Patsy.
+In addtion to the above dependencies, the GLM submodule relies on
+Patsy.
 
 [`scikits.sparse`](https://github.com/njsmith/scikits-sparse) enables sparse scaling matrices which are useful for large problems. Installation on Ubuntu is easy:
 

setup.py

@@ -27,7 +27,7 @@
                'Operating System :: OS Independent']
 
 install_reqs = ['numpy>=1.7.1', 'scipy>=0.12.0', 'matplotlib>=1.2.1',
-                'Theano<=0.7.1dev']
+                'Theano<=0.7.1dev', 'pandas>=0.15.0']
 
 test_reqs = ['nose']
 if sys.version_info[0] == 2:  # py3 has mock in stdlib