def test_ML(self):
ksi0 = -0.1
mu0 = 1000.0
sigma0 = 100.0
vals = []
ps_0 = []
ps_lm = []
ps_gml = []
ps_for_vals = []
for i in range(30):
p = random.random()
ps_for_vals.append(p)
vals.append(gev_cdf_inverse(p, sigma0, mu0, ksi0))
vals = np.array(vals)
xs = np.arange(800, 1500, 10)
sigma_lm, mu_lm, ksi_lm, dummy = gevfit.optimize_stationary_for_period(vals, use_lmoments = True)
sigma_gml, mu_gml, ksi_gml, dummy = gevfit.optimize_stationary_for_period(vals, use_lmoments = False)
for x in xs:
ps_0.append(gev_cdf(x, sigma0, mu0, ksi0))
ps_lm.append(gev_cdf(x, sigma_lm, mu_lm, ksi_lm))
ps_gml.append(gev_cdf(x, sigma_gml, mu_gml, ksi_gml))
print('LM: ', sigma_lm, mu_lm, ksi_lm)
print('GML: ', sigma_gml, mu_gml, ksi_gml)
print(ps_lm)
plt.title(r'$\sigma = %.2f , \mu = %.2f , \xi = %.2f$' % (sigma0, mu0, ksi0))
plt.plot(xs, ps_0, linewidth = 2, label = 'Theory')
plt.plot(xs, ps_lm, linewidth = 2, label = 'LM')
plt.plot(xs, ps_gml, linewidth = 2, label = 'GML')
plt.plot(vals, ps_for_vals, 'o', label = 'Sample')
plt.xlabel('$x$')
plt.ylabel('CDF')
plt.legend()
plt.show()
pass
def function_for_each_process(args):
sample_index, sampled_indices, extremes, \
return_periods, high_flow, positions = args
print('I work on sample %d' % sample_index)
result = {}
for the_period in return_periods:
result[the_period] = []
for pos in positions:
pars = gevfit.optimize_stationary_for_period(extremes[sampled_indices, pos],
high_flow = high_flow)
#calculate return levels for the current position and sample
for the_period in return_periods:
if high_flow:
ret_level = gevfit.get_high_ret_level_stationary(pars, the_period)
else:
ret_level = gevfit.get_low_ret_level_stationary(pars, the_period)
result[the_period].append(ret_level)
return result