def skewness(self):
ans = self.KL_thi(0) / (self.KL_sec(0))**1.5
return ans
def kurtosis(self):
ans = self.KL_for(0) / (self.KL_sec(0))**2 + 3
return ans
if __name__ == '__main__':
pf = portfolio_info()
pf.init_obligor()
df = pf.df
cbvpara = CBVmodel()
cbvpara.CBV3()
coca = cgf_calculation(pf, cbvpara)
skew_value = coca.skewness()
kurt_value = coca.kurtosis()
coca.KL_sec(0)
root = coca.QL_root()
ans_set1 = []
ans_set2 = []
spt_set = np.arange(3, 5, 0.05)
i = 92
for spt in spt_set:
ans_set1.append(coca.new_KL_fir(spt, i))
return contr_df.loc[:, ["varc_mean", "varc_std", "esc_mean", "esc_std"]]
if __name__ == '__main__':
pathdic = os.path.join(fileDir + '/MCmethod/MCResult/obligor/CBV2/')
tp = np.arange(0.01, 0.11, 0.01)
datap = pan.read_csv(os.path.join(pathdic, 'pmc_loss.csv'))
pmc_ans = pmc_result(datap, tp)
losshood = np.arange(4.1, 4.35, 0.05)
datais = pan.read_csv(os.path.join(pathdic, 'ismc_data.csv'))
ismc_ans = ismc_result(datais, losshood, SINGLE=False)
pf = portfolio_info()
pf.init_obligor()
cbvpara = CBVmodel()
cbvpara.CBV2()
coca = cgf_calculation(pf, cbvpara)
root = coca.QL_root()
tailprob_set = np.arange(0.001, 0.05, 0.001)
ans_set = pan.DataFrame(
columns=["var1", "var2", "var_check", "es1", "es2", "es3", "es_check"])
init = np.array(2.5, dtype="float")
for idx, i in enumerate(tailprob_set):
model = SPAcalculation(coca, est_spt=1.2, est_var=2.5, tailprob=i)
var1, spt1 = model.solver_tailprob_changeto_VaR_spt()
var2, spt2 = model.solver_tailprob_changeto_VaR_spt_2nd()
es1 = model.ES1()
es2 = model.ES2()
es3 = model.ES3()
es4 = minimize(model.check_function, init, method='Nelder-Mead')
sing_contribution += tp_hat2 * self.gamma[idx_common][
idx_sector]
# put into set
set_contribution.append(
(sing_exposure * sing_pd * sing_contribution / tp)[0])
return set_contribution
if __name__ == '__main__':
pf = portfolio_info()
pf.init_rcobligor1()
cbvpara = CBVmodel()
cbvpara.CBV1()
coca = cgf_calculation(pf, cbvpara)
root = coca.QL_root()
skew_value = coca.skewness()
kurt_value = coca.kurtosis()
coca.KL_sec(0)
print("the mean of the portfolio loss:", coca.Lmean)
print("the first order derivative of cgf at t=0:", coca.KL_fir(0))
print("minimum upper bound of t inside the root of cgf:", root)
ans_set = []
spt_set = np.arange(0.01, root - 0.005, 0.05)