def data_description(index, start, end):
returns = download_data.get_returns(index, start, end)
print('个数：', len(returns))
print('平均值:', np.mean(returns))
print('中位数:', np.median(returns))
print('上四分位数', sts.quantile(returns, p=0.25))
print('下四分位数', sts.quantile(returns, p=0.75))
#离散趋势的度量
print('最大值:', np.max(returns))
print('最小值:', np.min(returns))
print('极差:', np.max(returns) - np.min(returns))
print('四分位差',
sts.quantile(returns, p=0.75) - sts.quantile(returns, p=0.25))
print('标准差:', np.std(returns))
print('方差:', np.var(returns))
print('离散系数:', np.std(returns) / np.mean(returns))
#偏度与峰度的度量
print('偏度:', sts.skewness(returns))
print('峰度:', sts.kurtosis(returns))
print(st.kstest(returns, 'norm'))
length = len(returns)
sns.distplot(returns, bins=100, label='Empirical')
sns.plt.legend()
sns.plt.title('Empirical')
sns.plt.show()
def draw_lognorm(index, start, end):
returns = download_data.get_returns(index, start, end)
fit_data = st.norm(returns.mean(), returns.std()).rvs(10000000)
sns.distplot(fit_data, label='log_norm', hist=None)
length = len(returns)
sns.distplot(returns, bins=100, label='Empirical')
sns.plt.legend()
sns.plt.title('LogNorm')
sns.plt.show()
def draw_mixture(index, start, end):
returns = download_data.get_returns(index, start, end)
length = len(returns)
returns = returns.reshape((length, 1))
aic = [0 for i in range(4)]
bic = [0 for i in range(4)]
for i in range(2, 6):
gmm = GaussianMixture(n_components=i, tol=1e-6, covariance_type='full')
fgmm = gmm.fit(returns)
aic[i - 2] = fgmm.aic(returns)
bic[i - 2] = fgmm.bic(returns)
k1 = int(aic.index(min(aic))) + 2
k2 = int(bic.index(min(bic))) + 2
print(k1)
print(k2)
if k1 == k2:
gmm = GaussianMixture(n_components=k1,
tol=1e-6,
covariance_type='full')
fgmm = gmm.fit(returns)
sample1 = ((fgmm.sample(10000000))[0].reshape(1, 10000000))
sns.distplot(sample1, label='Gaussian', hist=None)
sns.distplot(returns, bins=100, label='Empirical')
sns.plt.legend()
sns.plt.title('GaussianMixture')
sns.plt.show()
else:
gmm1 = GaussianMixture(n_components=k1,
tol=1e-5,
covariance_type='full')
fgmm1 = gmm1.fit(returns)
gmm2 = GaussianMixture(n_components=k2,
tol=1e-5,
covariance_type='full')
fgmm2 = gmm2.fit(returns)
sample1 = ((fgmm1.sample(10000000))[0].reshape(1, 10000000))
sample2 = ((fgmm2.sample(10000000))[0].reshape(1, 10000000))
sns.distplot(sample1, label='Gaussian-AIC', hist=None)
sns.distplot(sample2, label='Gaussian-BIC', hist=None)
sns.distplot(returns, bins=100, label='Empirical')
sns.plt.legend()
sns.plt.title('GaussianMixture')
sns.plt.show()
import seaborn as sns
import numpy as np
import pandas as pd
from statsmodels.tsa.stattools import acf
from statsmodels.sandbox.stats.runs import runstest_1samp
import download_data
def get_log_return(file, length):
data = pd.read_csv(file, encoding='gbk')
returns = np.array(data[u'收盘价'] / data[u'前收盘价'])[-length:]
log_returns = np.log(returns)
return log_returns
SH = download_data.get_returns('sh000001', '20100101', '20170413')
SZ = download_data.get_returns('sz399001', '20100101', '20170413')
length1 = len(SZ)
HSI = get_log_return('HSI.CSV', length1)
DJONES = get_log_return('DJONES.CSV', length1)
SP500 = get_log_return('SP500.CSV', length1)
NASDAQ = get_log_return('NASDAQ.CSV', length1)
N225 = get_log_return('N225.CSV', length1)
def autocor_test(data):
lag = int((len(data))**0.5)
acf_result = acf(data, nlags=lag, qstat=True, alpha=0.05)
runstest_result = runstest_1samp(data, cutoff='mean')
plot_acf(data, lags=lag, alpha=0.05)