Testing mc_sample_path with np.random.seed

[oyamad]

Following the discussion in #147:

>>> import numpy as np
>>> prng = np.random.RandomState(42)
>>> u = prng.random_sample(size=10)
>>> u
array([ 0.37454012,  0.95071431,  0.73199394,  0.59865848,  0.15601864,
        0.15599452,  0.05808361,  0.86617615,  0.60111501,  0.70807258])

Let P = [[0.4, 0.6], [0.2, 0.8]], init = 0 (so u[0] is not used), and sample_size = 10. Given the values in u[1:], mc_sample_path should return [0, 1, 1, 1, 0, 0, 0, 1, 1, 1].

[coveralls]

Coverage increased (+0.02%) to 89.07% when pulling bdc6e67 on test_mc_sample_path into a9ee37a on master.

[jstac]

@oyamad I just commented on #147 without noticing this PR. What do you think about adding an optional argument to supply a seed to mc_sample_path?

To me it does seem desirable to allow users to supply seeds in a straightforward way.

[mmcky]

@oyamad Has this issue now been addressed with the inclusion of PR #154?

[oyamad]

@mmcky This is addressed by PR #166. See also my comment there.

[mmcky]

Great - thanks. I will close this PR then in preference or PR #166.