def inputs():
out = {}
disc_fac = 0.9999
num_states = 300
num_buses = 200
num_periods = 1000
scale = 0.001
init_dict = {
"groups": "group_4",
"binsize": 5000,
"model_specifications": {
"discount_factor": disc_fac,
"number_states": num_states,
"maint_cost_func": "linear",
"cost_scale": scale,
},
"optimizer": {"approach": "NFXP", "algorithm": "scipy_L-BFGS-B"},
"simulation": {
"discount_factor": disc_fac,
"seed": 123,
"buses": num_buses,
"periods": num_periods,
},
}
out["trans_base"] = np.loadtxt(TEST_FOLDER + "repl_test_trans.txt")
out["params_base"] = np.loadtxt(TEST_FOLDER + "repl_params_linear.txt")
trans_mat = create_transition_matrix(num_states, out["trans_base"])
costs = calc_obs_costs(num_states, lin_cost, out["params_base"], scale)
ev_known = calc_fixp(trans_mat, costs, disc_fac)[0]
df = simulate(init_dict["simulation"], ev_known, costs, trans_mat)
result_trans, result_fixp = estimate(init_dict, df)
out["trans_est"] = result_trans["x"]
out["params_est"] = result_fixp["x"]
out["status"] = result_fixp["status"]
return out
def test_regression_simulation(inputs):
init_dict = random_init(inputs)
# Draw parameter
param_1 = np.random.normal(17.5, 2)
param_2 = np.random.normal(21, 2)
param_3 = np.random.normal(-2, 0.1)
param_4 = np.random.normal(0.2, 0.1)
params = np.array([param_1, param_2, param_3, param_4])
disc_fac = init_dict["simulation"]["discount_factor"]
probs = np.array(init_dict["simulation"]["known_trans"])
num_states = 800
trans_mat = create_transition_matrix(num_states, probs)
costs = calc_obs_costs(num_states, cubic_costs, params, 0.00001)
ev = calc_fixp(trans_mat, costs, disc_fac)[0]
df = simulate(init_dict["simulation"], ev, costs, trans_mat)
v_disc = discount_utility(df, disc_fac)
assert_allclose(v_disc / ev[0], 1, rtol=1e-02)
def simulate_data(
seed,
disc_fac,
num_buses,
num_periods,
num_states,
cost_params,
trans_params,
cost_func,
scale,
):
"""
simulates a single data set with a given specification using the ``simulate``
function of ruspy.
Parameters
----------
seed : int
seed for the simulation function.
disc_fac : float
the discount factor in the Rust Model.
num_buses : int
the amount of buses that should be simulated.
num_periods : int
The number of periods that should be simulated for each bus.
num_states : int
the number of states for the which the mileage state is discretized.
cost_params : np.array
the cost parameters for which the data is simulated.
trans_params : np.array
the cost parameters for which the data is simulated..
cost_func : callable
the cost function that underlies the data generating process.
scale : float
the scale of the cost function.
Returns
-------
df : pd.DataFrame
Simulated data set for the given data generating process.
"""
init_dict = {
"simulation": {
"discount_factor": disc_fac,
"periods": num_periods,
"seed": seed,
"buses": num_buses,
},
}
costs = calc_obs_costs(num_states, cost_func, cost_params, scale)
trans_mat = create_transition_matrix(num_states, trans_params)
ev = calc_fixp(trans_mat, costs, disc_fac)[0]
df = simulate(init_dict["simulation"], ev, costs, trans_mat)
return df
def test_regression_simulation_reduced_data_discounted_utility(inputs_sim):
utility = simulate(
inputs_sim["init_dict"],
inputs_sim["ev"],
inputs_sim["costs"],
inputs_sim["trans_mat"],
reduced_data="discounted utility",
)
assert_almost_equal(utility, inputs_sim["v_disc"], decimal=4)
def test_regression_simulation_reduced_data_utility(inputs_sim):
utilities = simulate(
inputs_sim["init_dict"],
inputs_sim["ev"],
inputs_sim["costs"],
inputs_sim["trans_mat"],
reduced_data="utility",
)
assert_array_equal(
utilities,
inputs_sim["df"]["utilities"].to_numpy().reshape(
(inputs_sim["init_dict"]["buses"],
inputs_sim["init_dict"]["periods"])),
)
def test_regression_simulation(inputs):
init_dict = random_init(inputs)
df, unobs, utilities, num_states = simulate(init_dict["simulation"])
num_buses = init_dict["simulation"]["buses"]
num_periods = init_dict["simulation"]["periods"]
beta = init_dict["simulation"]["beta"]
params = np.array(init_dict["simulation"]["params"])
probs = np.array(init_dict["simulation"]["known probs"])
v_disc_ = [0.0, 0.0]
v_disc = discount_utility(
v_disc_, num_buses, num_periods, num_periods, utilities, beta
)
trans_mat = create_transition_matrix(num_states, probs)
costs = myopic_costs(num_states, lin_cost, params)
v_calc = calc_fixp(num_states, trans_mat, costs, beta)
un_ob_av = 0
for bus in range(num_buses):
un_ob_av += unobs[bus, 0, 0]
un_ob_av = un_ob_av / num_buses
assert_allclose(v_disc[1] / (v_calc[0] + un_ob_av), 1, rtol=1e-02)
def plot_convergence(init_dict):
beta = init_dict["simulation"]["beta"]
df, unobs, utilities, num_states = simulate(init_dict["simulation"])
costs = myopic_costs(num_states, lin_cost,
init_dict["simulation"]["params"])
trans_probs = np.array(init_dict["simulation"]["known probs"])
trans_mat = create_transition_matrix(num_states, trans_probs)
ev = calc_fixp(num_states, trans_mat, costs, beta)
num_buses = init_dict["simulation"]["buses"]
num_periods = init_dict["simulation"]["periods"]
gridsize = init_dict["plot"]["gridsize"]
num_points = int(num_periods / gridsize)
v_exp = np.full(num_points, calc_ev_0(ev, unobs, num_buses))
v_start = np.zeros(num_points)
v_disc = discount_utility(v_start, num_buses, gridsize, num_periods,
utilities, beta)
periods = np.arange(0, num_periods, gridsize)
ax = plt.figure(figsize=(14, 6))
ax1 = ax.add_subplot(111)
ax1.set_ylim([0, 1.3 * v_disc[-1]])
ax1.set_ylabel(r"Value at time 0")
ax1.set_xlabel(r"Periods")
ax1.plot(periods, v_disc, color="blue")
ax1.plot(periods, v_exp, color="orange")
plt.tight_layout()
os.makedirs("figures", exist_ok=True)
plt.savefig("figures/figure_1.png", dpi=300)
def inputs_sim(inputs):
out = {}
out["init_dict"] = init_dict = random_init(inputs)["simulation"]
# Draw parameter
param1 = np.random.normal(10.0, 2)
param2 = np.random.normal(2.3, 0.5)
params = np.array([param1, param2])
disc_fac = init_dict["discount_factor"]
probs = np.array(init_dict["known_trans"])
num_states = 300
out["trans_mat"] = trans_mat = create_transition_matrix(num_states, probs)
out["costs"] = costs = calc_obs_costs(num_states, lin_cost, params, 0.001)
out["ev"] = ev = calc_fixp(trans_mat, costs, disc_fac)[0]
out["df"] = simulate(
init_dict,
ev,
costs,
trans_mat,
)
return out
init_dict = yaml.safe_load(y)
beta = init_dict["simulation"]["beta"]
init_dict["simulation"]["states"] = 90
v_exp_known = []
v_exp_real = []
v_disc = []
roh_plot = []
for roh in np.arange(0, 4, 0.1):
roh_plot += [roh]
init_dict["simulation"]["roh"] = roh
worst_trans = get_worst_trans(init_dict["simulation"])
init_dict["simulation"]["real probs"] = worst_trans
df, unobs, utilities, num_states = simulate(init_dict["simulation"])
costs = myopic_costs(num_states, lin_cost, init_dict["simulation"]["params"])
num_buses = init_dict["simulation"]["buses"]
num_periods = init_dict["simulation"]["periods"]
gridsize = init_dict["plot"]["gridsize"]
real_trans_probs = np.array(init_dict["simulation"]["real probs"])
real_trans_mat = create_transition_matrix(num_states, real_trans_probs)
ev_real = calc_fixp(num_states, real_trans_mat, costs, beta)
known_trans_probs = np.array(init_dict["simulation"]["known probs"])
known_trans_mat = create_transition_matrix(num_states, known_trans_probs)
ev_known = calc_fixp(num_states, known_trans_mat, costs, beta)
def inputs():
out = {}
out["df"], out["unobs"], out["utilities"], num_states = simulate(
init_dict["simulation"]
)
return out