def test():
#%% Vasicek model
kappa, mu, sigma = 1.5, .5, .1
theta_true = np.array([kappa, mu, sigma])
x0, T, h, M, S = mu, 200, 1., 100, 1e3
N = int(float(T) / h)
vasicek = Vasicek()
vasicek.simulate(x0, theta_true, h, M, N, S)
vasicek.plot_trajectories(3)
vasicek.plot_final_distr()
r = vasicek.paths[:, 0]
compare_estimators(vasicek, r, theta_true)
def test():
#%% Vasicek model
kappa, mu, sigma = 1.5, .5, .1
theta_true = np.array([kappa, mu, sigma])
x0, T, h, M, S = mu, 200, 1., 100, 1e3
N = int(float(T) / h)
vasicek = Vasicek()
vasicek.simulate(x0, theta_true, h, M, N, S)
vasicek.plot_trajectories(3)
vasicek.plot_final_distr()
r = vasicek.paths[:,0]
compare_estimators(vasicek, r, theta_true)
def test():
#%% Cox-Ingersoll-Ross model
kappa, mu, sigma = 1, .2, .2
theta_true = np.array([kappa, mu, sigma])
x0, T, h, M, S = mu, 200, 1., 100, 1e2
N = int(float(T) / h)
cir = CIR()
cir.simulate(x0, theta_true, h, M, N, S)
cir.is_valid(theta_true)
cir.plot_trajectories(3)
cir.plot_final_distr()
r = cir.paths[:,0]
compare_estimators(cir, r, theta_true)
def test():
#%% GBM model
# At this point we can initialize a model object using the class GBM.
mu, sigma = .05, .1
theta_true = np.array([mu, sigma])
gbm = GBM()
x0, T, h, M, S = mu, 200, 1., 100, 3
N = int(float(T) / h)
gbm.simulate(x0, theta_true, h, M, N, S)
# Calling another method creates a simple plot.
gbm.plot_trajectories(3)
# Use the first path for furtehr estimation.
logS = gbm.paths[:, 0]
compare_estimators(gbm, logS, theta_true)