def np_hess(log_like, params, method, log_like_kwargs):
"""We wrote this function because we did not want to touch the
differentiation files.
"""
numpy_hessian = hessian(log_like,
params,
method,
func_kwargs=log_like_kwargs)
return numpy_hessian.to_numpy()
def test_hessian(statsmodels_fixtures, method, extrapolation):
fix = statsmodels_fixtures
calculated = hessian(
logit_loglike,
fix["params"],
method=method,
extrapolation=extrapolation,
func_kwargs={
"y": fix["y"],
"x": fix["x"]
},
)
expected = fix["hessian"]
assert_frame_equal(calculated, expected)
def params_hess(params, df, beta, maint_func, repl_4=False):
"""Calculates the hessian of the cost parameters."""
transition_results = estimate_transitions(df, repl_4=repl_4)
ll_trans = transition_results["fun"]
states = df.loc[:, "state"].to_numpy()
num_obs = df.shape[0]
if repl_4:
num_states = 90
else:
num_states = int(1.2 * np.max(states))
trans_mat = create_transition_matrix(num_states,
np.array(transition_results["x"]))
state_mat = create_state_matrix(states, num_states, num_obs)
endog = df.loc[:, "decision"].to_numpy()
decision_mat = np.vstack(((1 - endog), endog))
params_df = pd.DataFrame(index=["RC", "theta_1_1"],
columns=["value"],
data=params)
wrap_func = create_wrap_func(maint_func, num_states, trans_mat, state_mat,
decision_mat, beta, ll_trans)
return hessian(wrap_func, params_df)