def unitroot_test(series):
# Basic statistic
plt.figure()
plt.plot(series)
plot_pacf(series)
# ADF test
# AIC & BIC from lags 12 to 1
print('$p$ & AIC & BIC \\\\')
max_lags = 12
for lags in (max_lags - i for i in range(max_lags)):
ar_model = AutoReg(series, lags, 'n')
res = ar_model.fit()
print(f'{lags} & {round(res.aic, 3)} & {round(res.bic, 3)} \\\\')
# Best lags by `ar_select_order`
sel = ar_select_order(series, max_lags, trend='n')
lags = sel.ar_lags[-1]
print(f'Lags selection: {sel.ar_lags}')
# Start ADF test
adf = ADF(series, lags)
print(adf.summary())
# PP test
pp_tau = PhillipsPerron(series, 3, test_type='tau') # q = 3
pp_rho = PhillipsPerron(series, 3, test_type='rho') # q = 3
print(pp_tau.summary())
print(pp_rho.summary())
def AugmentedDickeyFullerTest(data, printResults=True, trend=None, lags=None):
options_Trend = trend if trend != None else {'nc','c','ct','ctt'}
options_Lags = lags if lags != None else {0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24}
#options_LagMethod = lagMethod if lagMethod != None else {'AIC', 'BIC', 't-stat', None}
results = dict()
for column in data.columns:
print("Augmented Dickey Fuller test for column: " + column)
results_Trend = dict()
for option_Trend in options_Trend:
results_Lag = dict()
for option_Lag in options_Lags:
result = ADF(data[column].dropna(), trend=option_Trend, lags=option_Lag)
if printResults:
result.summary()
results_Lag[option_Lag] = result
results_Trend[option_Trend] = results_Lag
results[column] = results_Trend
return results
def UnitRootTest(ret, isdefaultmethod=True, isprintsummary=False):
"""
進行 Augmented Dickey-Fuller (ADF) 單根檢定
若接受虛無假設則表示有單根，序列為非定態 (non-stationary)，回傳值為False
若拒絕虛無假設則表示無單根，序列為弱定態 (weakly stationary)，回傳值為True
"""
# 除非進行除錯，否則建議參數使用預設值
if isdefaultmethod: # default & recommended in the current version
results = ADF(ret, lags = None, max_lags = None, \
trend = 'c', method = "AIC")
if isprintsummary:
print(results.summary().as_text())
if results.pvalue < PVal: # ADF測試對應p-value小於5%
if isprintsummary:
print("- - - - - - - - - - - - - - - - - - - - - - - - - -")
print("ADF stats p-value < %.2f -> weakly stationary!" % PVal)
print("- - - - - - - - - - - - - - - - - - - - - - - - - -")
return True
else:
if isprintsummary:
print("- - - - - - - - - - - - - - - - - - - - - - - - - -")
print("ADF stats p-value > %.2f -> non-stationary!" % PVal)
print("- - - - - - - - - - - - - - - - - - - - - - - - - -")
return False
else: # future preference
results = stattools.adfuller(ret, maxlag = None, \
regression = 'c', autolag = "AIC", \
store = False, regresults = False)
# returns: adfstat, pvalue, usedlag, nobs, critvalues(, icbest)
if isprintsummary:
print("ADF test statistics: ", results[0])
print("MacKinnon approximated p-value: ", results[1])
print("# of lags used: ", results[2])
print("# of obs. used for ADF test: ", results[3])
print("Critical values for p-value = 0.01, 0.05, and 0.10: ")
print(results[4]["1%"])
print(results[4]["5%"])
print(results[4]["10%"])
print("The best information criterion (min. AIC): ", results[5])
if results[1] < PVal: # ADF測試對應p-value小於5%
if isprintsummary:
print("- - - - - - - - - - - - - - - - - - - - - - - - - -")
print("ADF stats p-value < %.2f -> weakly stationary!" % PVal)
print("- - - - - - - - - - - - - - - - - - - - - - - - - -")
return True
else:
if isprintsummary:
print("- - - - - - - - - - - - - - - - - - - - - - - - - -")
print("ADF stats p-value > %.2f -> non-stationary!" % PVal)
print("- - - - - - - - - - - - - - - - - - - - - - - - - -")
return False
def ADF_test(self, df_ts, lags=None):
if lags == 'None':
try:
adf = ADF(df_ts)
except:
adf = ADF(df_ts.dropna())
else:
try:
adf = ADF(df_ts)
except:
adf = ADF(df_ts.dropna())
adf.lags = lags
print(adf.summary().as_text())
return adf
def test():
df = merge_prices(get_price('300033'), get_price('300059'))
train_df, test_df = split_by_date(df, '2015-12-31')
x, y = train_df.close_x, train_df.close_y
# print(ADF(x).summary())
# print(ADF(y).summary())
# print(ADF(np.diff(x)).summary())
# print(ADF(np.diff(y)).summary())
modol = sm.OLS(y, sm.add_constant(x))
result = modol.fit()
# print(result.summary())
residual = result.resid
adf = ADF(residual)
print(adf.summary())
def time_series(self): # 时间序列
rate1 = self.rate
# 计算自相关系数
acfs = stattools.acf(rate1)
# 计算偏自相关系数
pacfs = stattools.pacf(rate1)
# 绘制自相关系数图
plot_acf(rate1, use_vlines=True, lags=30)
# 绘制偏自相关系数图
plot_pacf(rate1, use_vlines=True, lags=30)
# 平稳性 1 看时序图 2 看自相关和偏自相关 3 单位根检验DF ADF PP检验
# ADF检验
adfrate = ADF(rate1)
print(adfrate.summary().as_text())
pass
def ADF_test(df_ts, lags=None):
"""
ADF from arch
formula:
xt-xt-1 ~ b0 + (b1-1)*xt-1 + e
test if b1-1 == 0 ~ DF statistics
:param df_ts:
:param lags:
:return:
"""
if lags == 'None':
try:
adf = ADF(df_ts)
except:
adf = ADF(df_ts.dropna())
else:
try:
adf = ADF(df_ts)
except:
adf = ADF(df_ts.dropna())
adf.lags = lags
print(adf.summary().as_text())
return adf
@author: Liu
"""
import numpy as np
import pandas as pd
import pandas_datareader as web
import matplotlib.pyplot as plt
import datetime
start=datetime.datetime(2018,1,1)
end=datetime.datetime(2019,6,30)
nasdaq=web.DataReader('^IXIC','yahoo',start,end)
nasdaq=nasdaq.dropna()
nasdaq=pd.DataFrame(nasdaq['Adj Close'].values,index=pd.to_datetime(nasdaq.index),columns=['Price'])
nasdaq=(nasdaq-nasdaq.shift(1))/nasdaq.shift(1)
nasdaq=nasdaq.dropna()
nasdaq.columns=['Return']
nasdaq.tail(3)
traindata=nasdaq[:-3]
from arch.unitroot import ADF
result=ADF(traindata.Return,max_lags=10)
print(result.summary().as_text())
from statsmodels.tsa import stattools
LjungBox=stattools.q_stat(stattools.acf(traindata)[1:12],len(traindata))
LjungBox[1][-1]
import statsmodels.graphics.tsaplots as ts
ts.plot_acf(traindata,use_vlines=True, lags=30)
if priceX is None or priceY is None:
print('缺少价格序列.')
returnX = (priceX - priceX.shift(1)) / priceX.shift(1)[1:]
returnY = (priceY - priceY.shift(1)) / priceY.shift(1)[1:]
standardX = (returnX + 1).cumprod()
standardY = (returnY + 1).cumprod()
SSD = np.sum((standardX - standardY)**2)
return SSD
dis = SSD(PAf, PBf)
print(dis)
PAflog = np.log(PAf)
adfA = ADF(PAflog)
print(adfA.summary().as_text())
retA = PAflog.diff()[1:]
adfretA = ADF(retA)
print(adfretA.summary().as_text())
PBflog = np.log(PBf)
adfB = ADF(PBflog)
print(adfB.summary().as_text())
retB = PBflog.diff()[1:]
adfretB = ADF(retB)
print(adfretB.summary().as_text())
PAflog.plot(label='601988', style='--')
PBflog.plot(label='600000', style='-')
plt.legend(loc='upper left')
plt.title('中国银行与浦发银行的对数价格时序图')
# 上证综指的平稳性
SHClose = SHindex.Clsindex
fig = plt.figure()
SHClose.plot()
fig.savefig("stabibly.png")
fig = plt.figure()
SHRet.plot()
fig.savefig("SHret.png")
fig = plt.figure()
fig = plot_acf(SHClose, use_vlines=True, lags=30)
fig.savefig("SHRet_acf.png")
# 单位根检验
adfSHRet = ADF(SHRet)
print(adfSHRet.summary().as_text())
adfSHClose = ADF(SHClose)
print(adfSHClose.summary().as_text())
# 白噪声
whiteNoise = np.random.standard_normal(size=500)
fig = plt.figure()
plt.plot(whiteNoise, c="b")
fig.savefig("whiteNoise.png")
# 上证综指的白噪声检测
LB1 = stattools.q_stat(stattools.acf(SHRet)[1:13], len(SHRet))
print(LB1)
print(LB1[1][-1])
LB2 = stattools.q_stat(stattools.acf(SHClose)[1:13], len(SHClose))
print(LB2[1][-1])
def do_pair_trading_org(self):
sh = pd.read_csv('./data/sh50p.csv', index_col='Trddt')
sh.index = pd.to_datetime(sh.index)
#配对交易实测
#提取形成期数据
formStart = '2014-01-01'
formEnd = '2015-01-01'
PA = sh['601988']
PB = sh['600000']
PAf = PA[formStart:formEnd]
PBf = PB[formStart:formEnd]
#形成期协整关系检验
#一阶单整检验
log_PAf = np.log(PAf)
adfA = ADF(log_PAf)
print(adfA.summary().as_text())
adfAd = ADF(log_PAf.diff()[1:])
print(adfAd.summary().as_text())
# B股票平稳性检查
log_PBf = np.log(PBf)
adfB = ADF(log_PBf)
print(adfB.summary().as_text())
adfBd = ADF(log_PBf.diff()[1:])
print(adfBd.summary().as_text())
#
#协整关系检验
model = sm.OLS(log_PBf, sm.add_constant(log_PAf)).fit()
print('model:\r\n{0}'.format(model.summary()))
alpha = model.params[0]
print('alpha={0};'.format(alpha))
beta = model.params[1]
print('beta={0}'.format(beta))
#残差单位根检验
spreadf = log_PBf - beta * log_PAf - alpha
adfSpread = ADF(spreadf)
print(adfSpread.summary().as_text())
#
mu = np.mean(spreadf)
sd = np.std(spreadf)
#
#设定交易期
tradeStart = '2015-01-01'
tradeEnd = '2015-06-30'
PAt = PA[tradeStart:tradeEnd]
PBt = PB[tradeStart:tradeEnd]
CoSpreadT = np.log(PBt) - beta * np.log(PAt) - alpha
print('CoSpreadT: {0};'.format(CoSpreadT.describe()))
plt.rcParams['font.family'] = 'sans-serif'
plt.rcParams['font.sans-serif'] = ['SimHei']
plt.rcParams['axes.unicode_minus'] = False
CoSpreadT.plot()
plt.title('交易期价差序列(协整配对)')
plt.axhline(y=mu, color='black')
plt.axhline(y=mu + 0.2 * sd, color='blue', ls='-', lw=2)
plt.axhline(y=mu - 0.2 * sd, color='blue', ls='-', lw=2)
plt.axhline(y=mu + 1.5 * sd, color='green', ls='--', lw=2.5)
plt.axhline(y=mu - 1.5 * sd, color='green', ls='--', lw=2.5)
plt.axhline(y=mu + 2.5 * sd, color='red', ls='-.', lw=3)
plt.axhline(y=mu - 2.5 * sd, color='red', ls='-.', lw=3)
plt.show()
#
level = (float('-inf'), mu - 2.5 * sd, mu - 1.5 * sd, mu - 0.2 * sd,
mu + 0.2 * sd, mu + 1.5 * sd, mu + 2.5 * sd, float('inf'))
print('level: {0}={1}'.format(type(level), level))
#
prcLevel = pd.cut(CoSpreadT, level, labels=False) - 3
print('prcLevel: {0}'.format(prcLevel.head()))
signal = self.TradeSig(prcLevel)
print('signal: {0}={1}'.format(type(signal), signal))
# position
position = [signal[0]]
ns = len(signal)
for i in range(1, ns):
position.append(position[-1])
if signal[i] == 1:
position[i] = 1
elif signal[i] == -2:
position[i] = -1
elif signal[i] == -1 and position[i - 1] == 1:
position[i] = 0
elif signal[i] == 2 and position[i - 1] == -1:
position[i] = 0
elif signal[i] == 3:
position[i] = 0
elif signal[i] == -3:
position[i] = 0
position = pd.Series(position, index=CoSpreadT.index)
print('position: {0}'.format(position.tail()))
#
account = self.TradeSim(alpha, beta, PAt, PBt, position)
print('account: {0}'.format(account.tail()))
#
account.iloc[:, [0, 1, 4]].plot(style=['--', '-', ':'])
plt.title('配对交易账户')
plt.show()
def test_adf_lags_10(self):
adf = ADF(self.inflation, lags=10)
assert_almost_equal(adf.stat, -2.28375, DECIMAL_4)
adf.summary()
garch_plot1(data['Close'])
lh = np.log(data / data.shift(1)).dropna() # d 1
garch_plot1(lh['Close'])
print('Lean Hogs Future skewness is {}'.format(lh.skew(axis=0)[0]))
print('Lean Hogs Future kurtosis is {}'.format(lh.kurtosis(axis=0)[0]))
sns.distplot(lh['Close'], color='blue') #density plot
plt.title('1986–2018 Lean Hogs Future return frequency')
plt.xlabel('Possible range of data values')
# Pull up summary statistics
print(lh.describe())
adf = ADF(lh['Close'])
print(adf.summary().as_text())
kpss = KPSS(lh['Close'])
print(kpss.summary().as_text())
dfgls = DFGLS(lh['Close'])
print(dfgls.summary().as_text())
pp = PhillipsPerron(lh['Close'])
print(pp.summary().as_text())
za = ZivotAndrews(lh['Close'])
print(za.summary().as_text())
vr = VarianceRatio(lh['Close'], 12)
print(vr.summary().as_text())
from arch import arch_model
X = 100 * lh
plot_pacf(taiexRet,use_vlines=True,lags=30)
close.plot()
plt.title('2014-2015年加權股價指數收盤指數時序圖 ')
taiexRet.plot()
plt.title('2014-2015年加權股價指數收益率指數時序圖')
plot_acf(taiexRet,use_vlines=True,lags=30)
plot_pacf(taiexRet,use_vlines=True,lags=30)
plot_acf(close,use_vlines=True,lags=30)
adf_taiexRet=ADF(taiexRet)
print(adf_taiexRet.summary().as_text())
adfclose=ADF(close)
print(adfclose.summary().as_text())
#生成純隨機序列
whiteNoise=np.random.standard_normal(size=500)
#繪制該序列圖
plt.plot(whiteNoise,c='b')
plt.title('White Noise')
LjungBox1=stattools.q_stat(stattools.acf(taiexRet)[1:13],len(taiexRet))
LjungBox1
LjungBox1[1][-1]
#table2.to_csv('Table_2.csv')
### Table 3
#JC1.iloc[:, 2] = JC1.iloc[:, 2]*(-1)
x = JC1[['growth', 'VIX', 'BDI']]
x = sm.add_constant(x)
y = JC1['ore_price']
model_t3 = sm.OLS(y, x).fit(cov_type='HC1')
model_t3.summary()
residual_t3 = model_t3.resid
adf_resi_t3 = ADF(residual_t3)
adf_resi_t3.summary()
# Table for coefficient
table3_1 = pd.DataFrame(
index=['theta_0', 'theta_1', 'theta_2', 'theta_3'],
columns=['Estimated Coefficient', 'Standard Error', 't-stats', 'p-Value'])
t3_1_cols = table3_1.columns.tolist()
table3_1[t3_1_cols[0]] = model_t3.params.tolist()
table3_1[t3_1_cols[1]] = model_t3.bse.tolist()
table3_1[t3_1_cols[2]] = model_t3.tvalues.tolist()
table3_1[t3_1_cols[3]] = model_t3.pvalues.tolist()
# Table for F-stat, R^2, ADF
table3_2 = pd.DataFrame(index=[
'F-statistic', 'F-stat-pvalue', 'R^2', 'ADF-statistics Residual Value',
baiyun = pd.read_csv('data/baiyun.csv')
baiyun.index = pd.to_datetime(baiyun.Date)
# baiyun = baiyun.sort_index();
bclose = baiyun.Close
# print(bclose)
logReturn_all = pd.Series(np.log(bclose)).diff().dropna()
logReturn1 = logReturn_all[-10:]
logReturn = logReturn_all[:-10]
print(logReturn_all.shape, logReturn.shape, logReturn1.shape)
print(logReturn1)
# logReturn.plot()
adfReutn = ADF(logReturn, lags=6)
print(adfReutn.summary().as_text())
# plot_acf(logReturn, lags=20)
# plot_pacf(logReturn, lags=20)
model1 = arima_model.ARIMA(logReturn.values, order=(2, 0, 0)).fit()
model2 = arima_model.ARIMA(logReturn.values, order=(0, 0, 2)).fit()
print("model1.aic=%f \t model2.aic=%f" % (model1.aic, model2.aic))
print(model2.sigma2)
strresid = model2.resid / math.sqrt(model2.sigma2)
# strresid.plot()
plt.subplot(211)
plt.plot(strresid)
# plot_acf(strresid , lags=20)
LinjiuBox = stattools.q_stat(stattools.acf(strresid)[1:13], len(strresid))
# # plot_acf(ibm, use_vlines=True, lags=30)
# plot_acf(ibm, use_vlines=True, lags=20)
# adfIBM = ADF(ibm)
# print(adfIBM.summary().as_text())
# LinjiuBox = stattools.q_stat(stattools.acf(ibm)[1:13], len(ibm))
# print(LinjiuBox[1][-1])
# 第6题
# ge = CRSP.iloc[:,3]
#
# ge.plot()
# plot_acf(ge, use_vlines=True, lags=20)
# LinjiuBox = stattools.q_stat(stattools.acf(ge)[1:2], len(ge))
# print('lag=2',LinjiuBox[1][-1])
# LinjiuBox = stattools.q_stat(stattools.acf(ge)[1:9], len(ge))
# print('lag=9',LinjiuBox[1][-1])
# 第7题
r500 = SP500.r500
print(r500.head())
plt.subplot(221)
r500.plot(subplots=True)
plot_acf(r500, lags=20)
plot_pacf(r500, lags=20)
adfIBM = ADF(r500)
print(adfIBM.summary().as_text())
plt.show()
shindex = index['close'] shindex = shindex.diff(-1)/shindex.shift(-1) shindex.index = pd.to_datetime(shindex.index) shindex = shindex.dropna() acf = stattools.acf(shindex) pacf = stattools.pacf(shindex) plot_acf(shindex,use_vlines=True,lags=30) plot_pacf(shindex,use_vlines=True,lags=30) shindex.plot() adfshindex = ADF(shindex) print(adfshindex.summary().as_text()) whitenoise = np.random.standard_normal(500) plt.plot(whitenoise,c = 'b') cpi = ts.get_cpi() cpi.index = pd.to_datetime(cpi['month']) cpi = cpi['cpi'] cpitrain = cpi['2016-01-01':'2000-01-01'] cpitrain.plot() #是否平稳 print(ADF(cpitrain,max_lags=10).summary().as_text()) #是否白噪声 ljb0 = stattools.q_stat(stattools.acf(cpitrain)[1:12],len(cpitrain)) ljb0[1][-1]
def many_paras():
"""
p-value本质是控制假阳性率（False positive rate，FPR）
q-value 控制的是FDR (false discovery rate)
Q-statistic: Qlb=T*(T+2)*sigma(j=1,p)(rj^2/(T-j)) rj残差序列，j阶自相关系数，T观测值的个数，p滞后阶数。
FDR = E（V/R） 错误发现次数V，总的拒绝次数R
acf: 自相关系数 -- y(t)= a0 + a1*y(t-1) + epsilon
p(x(i)|x(i-h)) :sigma(i=1,n-h) ((x(i)-mu)*(x(i+h)-mu)/sigma(i=1,n) ((x(i)-mu)^2))
pacf: 偏自相关系数，k-1个时间滞后 作为已知，只求k -- y(t)= a0 + a1*y(t-1) + ... a1*y(t-k) + epsilon
p(x(i)..x(i-k)|x(i-1)x(i-k+1)) :
ARMA(p，q): AR代表p阶自回归过程，MA代表q阶移动平均过程
ARIMA模型是在ARMA模型的基础上多了差分的操作。
ADF: 白噪声随机干扰项的一阶自回归过程。用单位根 检验，存在就是非平稳。y(t)= mu + fi*y(t-1) + epsilon。p阶要求 p个根的和小于1。
Sma: 移动平均
wma: 加权移动平均
ema: 指数移动平均
ewma: 指数加权移动平均
OBV: On Balance Volume, 多空比率净额= [（收盘价－最低价）－（最高价-收盘价）] ÷（ 最高价－最低价）×V
:return:
"""
# 1. 计算自相关系数
acfs = stattools.acf(SHRet)
# 绘制自相关系数图
plot_acf(SHRet, use_vlines=True, lags=30)
# 2. 计算偏自相关系数
pacfs = stattools.pacf(SHRet)
plot_pacf(SHRet, use_vlines=True, lags=30)
# 3. 进行ADF单位根检验，并查看结果；
adfSHRet = ADF(SHRet)
print(adfSHRet.summary().as_text())
# 4. Q 统计
LjungBox1 = stattools.q_stat(stattools.acf(SHRet)[1:13], len(SHRet))
print(LjungBox1)
# 5. lag即为上述检验表达式中的m，在这里我们选择检验12阶的自相关系数。
LjungBox = stattools.q_stat(stattools.acf(CPItrain)[1:12], len(CPItrain))
# order表示建立的模型的阶数，c(1,0,1)表示建立的是ARMA(1,1)模型；
# 中间的数字0表示使用原始的、未进行过差分（差分次数为0）的数据；
model1 = arima_model.ARIMA(CPItrain, order=(1, 0, 1)).fit()
model1.summary()
model1.conf_int()
# 6. 绘制时间序列模拟的诊断图
stdresid = model1.resid / math.sqrt(model1.sigma2)
plt.plot(stdresid)
plot_acf(stdresid, lags=20)
LjungBox = stattools.q_stat(stattools.acf(stdresid)[1:13], len(stdresid))
print(LjungBox[1][-1])
print(model1.forecast(3)[0])
# 7. Autoregressive conditional heteroskedasticity model 自回归条件异方差模型
# y(t)=b*x(t)+epsilon(t)
# epsilon(t)^2=a0+a1*epsilon(t-1)^2+a2*epsilon(t-2)^2+n(t)
# \sigma_t^{2}=\omega+\sum_{i=1}^{p}\alpha_{i}\epsilon_{t-i}^{2}
# n(t)独立同分布 期望为0，var(n^2)=r^2
am = arch_model(SHret)
model = am.fit(update_freq=0)
print(model.summary())
# 8. 对子 的 处理
pt = TradeTool()
SSD = pt.SSD(priceAf, priceBf)
SSDspread = pt.SSDSpread(priceAf, priceBf)
SSDspread.describe()
coefficients = pt.cointegration(priceAf, priceBf)
CoSpreadF = pt.CointegrationSpread(priceA, priceB, formPeriod, formPeriod)
CoSpreadTr = pt.CointegrationSpread(priceA, priceB, formPeriod,
tradePeriod)
CoSpreadTr.describe()
bound = pt.calBound(priceA, priceB, 'Cointegration', formPeriod, width=1.2)
# 9. 配对 选点
trtl = TradeTool()
account = trtl.TradeSimPair(PAt, PBt, position)
# 10. momentum function
et = ElementTool()
et.momentum(Close, 5).tail(n=5)
momen35 = et.momentum(Close, 35)
signal = []
for i in momen35:
if i > 0:
signal.append(1)
else:
signal.append(-1)
signal = pd.Series(signal, index=momen35.index)
signal.head()
tradeSig = signal.shift(1)
ret = Close / Close.shift(1) - 1
# ret=ret['2014-02-20':]
# ret.head(n=3)
Mom35Ret = ret * (signal.shift(1))
Mom35Ret[0:5]
real_Mom35Ret = Mom35Ret[Mom35Ret != 0]
real_ret = ret[ret != 0]
Rsi12 = et.rsi(BOCMclp, 12)
# 策略
rsi6 = et.rsi(BOCMclp, 6)
rsi24 = et.rsi(BOCMclp, 24)
# rsi6捕捉买卖点
Sig1 = []
for i in rsi6:
if i > 80:
Sig1.append(-1)
elif i < 20:
Sig1.append(1)
else:
Sig1.append(0)
date1 = rsi6.index
Signal1 = pd.Series(Sig1, index=date1)
Signal1[Signal1 == 1].head(n=3)
Signal1[Signal1 == -1].head(n=3)
Signal2 = pd.Series(0, index=rsi24.index)
lagrsi6 = rsi6.shift(1)
lagrsi24 = rsi24.shift(1)
for i in rsi24.index:
if (rsi6[i] > rsi24[i]) & (lagrsi6[i] < lagrsi24[i]):
Signal2[i] = 1
elif (rsi6[i] < rsi24[i]) & (lagrsi6[i] > lagrsi24[i]):
Signal2[i] = -1
signal = Signal1 + Signal2
signal[signal >= 1] = 1
signal[signal <= -1] = -1
signal = signal.dropna()
tradSig = signal.shift(1)
tt = TradeTool()
BuyOnly = tt.strategy_analy(buy, ret)
SellOnly = tt.strategy_analy(sell, ret)
Trade = tt.strategy_analy(tradSig, ret)
Test = pd.DataFrame({
"BuyOnly": BuyOnly,
"SellOnly": SellOnly,
"Trade": Trade
})
# 累计收益率
cumStock = np.cumprod(1 + ret) - 1
cumTrade = np.cumprod(1 + tradeRet) - 1
# 12. 移动平均线
sma5 = et.smaCal(Close, 5)
# 12. 加权移动平均线
wma5 = et.wmaCal(Close, w)
# 12. 指数移动平均线
Ema = et.emaCal(Close, period)
print(Ema)
# 12. 指数加权移动平均线
Ewma = et.ewmaCal(Close, 5, 0.2)
# 13. 布林带
UnicomBBands = et.bbands(Close, 20, 2)
print(UnicomBBands)
multiplier = [1, 1.65, 1.96, 2, 2.58]
price2010 = Close['2010-01-04':'2010-12-31']
tt.CalBollRisk(price2010, multiplier)
# 14. 性能
btt = BackTestTool()
Performance1 = btt.perform(Close, tradSignal1)
print(Performance1)
# 15. 交易, 回测
KDtrade = btt.trade(KDSignal, close)
btt.backtest(KDtrade.Ret, KDtrade.KDtradeRet)
# 16. 上下突破
KDupbreak = et.upbreak(KValue, DValue) * 1
KDupbreak[KDupbreak == 1].head()
KDdownbreak = et.downbreak(KValue, DValue) * 1
KDdownbreak[KDdownbreak == 1].head()
# "金叉"与"死叉"交易策略绩效表现
btt.backtest(KDbreak.Ret, KDbreak.KDbreakRet)
# 17. 成交量指标
cumUpVol = et.VOblock(UpVol)
cumDownVol = et.VOblock(DownVol)
ALLVol = np.array([cumUpVol, cumDownVol]).transpose()
# 18. 判断持有
hold = tt.judge_hold(trade)
# 19. 单交易
TradeAccount = tt.TradeSim(close, hold)
print(TradeAccount)
Skew: -0.206 Prob(JB): 0.0558 Kurtosis: 2.486 Cond. No. 148. ============================================================================== Warnings: [1] Standard Errors assume that the covariance matrix of the errors is correctly specified. """ beta = result.params[1] # 残差 spread = PAlog - beta * ZSlog print(spread.head()) spread.plot() # 残差的均值为0,tread设置为c adfspread = ADF(spread, trend='c') print(adfspread.summary().as_text()) """ Name: close_price, dtype: float64 Augmented Dickey-Fuller Results ===================================== Test Statistic -1.429 P-value 0.568 Lags 0 ------------------------------------- Trend: Constant Critical Values: -3.45 (1%), -2.87 (5%), -2.57 (10%) Null Hypothesis: The process contains a unit root. Alternative Hypothesis: The process is weakly stationary. 不能拒绝原假设，误差太大了
numbers.plot()
plt.show()
plot_acf(numbers,lags=20)
from statsmodels.tsa import stattools
stattools.arma_order_select_ic(numbers.values,max_ma=4)
#7.
zgsy=pd.read_csv('Data/Part4/003/zgsy.csv')
clprice=zgsy.iloc[:,4]
clprice.plot()
plot_acf(clprice,lags=20)
from arch.unitroot import ADF
adf=ADF(clprice,lags=6)
print(adf.summary().as_text())
logReturn=pd.Series((np.log(clprice))).diff().dropna()
logReturn.plot()
adf=ADF(logReturn,lags=6)
print(adf.summary().as_text())
plot_acf(logReturn,lags=20)
plot_pacf(logReturn,lags=20)
from statsmodels.tsa import arima_model
model1=arima_model.ARIMA(logReturn.values,order=(0,0,1)).fit()
model1.summary()
model2=arima_model.ARIMA(logReturn.values,order=(1,0,0)).fit()
from sklearn.metrics import mean_squared_error as mse
from sklearn.metrics import r2_score as r2
from arch.unitroot import ADF
def parser(x):
return datetime.strptime(x,'%d/%m/%Y')
data = pd.read_csv('Bitcoin-Train.csv',index_col=0,parse_dates= [0],date_parser = parser)
test_data = pd.read_csv('Bitcoin-Test.csv',index_col=0,parse_dates = [0],date_parser = parser)
# Augmented Dickey Fuller Test for Stationarity
adf_test = ADF(data,regression = 'c')
adf_test.lags = 2
adf.trend = 'c'
print(adf_test.summary().as_text())
plt.plot(data)
# ACF and PACF of Raw data
plot_acf(data)
plot_pacf(data)
### Stationarity Conversion ###
data_diff = data.diff(periods = 1)
data_diff = data_diff[1:]
### ACF and PACF for differenced Time Series ###
plot_acf(data_diff)
plot_pacf(data_diff)
def test_adf_lags_10(self):
adf = ADF(self.inflation, lags=10)
assert_almost_equal(adf.stat, -2.28375, DECIMAL_4)
adf.summary()
def series(self):
rate = self.rate_log
fig = plt.figure(figsize=(20, 10), dpi=80, facecolor=[199 / 255, 238 / 255, 206 / 255], edgecolor='g')
fig.subplots_adjust(left=0.05, right=0.95, top=0.95, bottom=0.15)
ax = fig.add_subplot(1, 1, 1)
ax.set_facecolor([199 / 255, 238 / 255, 206 / 255])
min_ticks = math.floor(len(rate) / 30)
ax.set_xlabel('time')
ax.set_ylabel('price')
xtick = []
for i in range(0, len(rate), min_ticks):
xtick.append(i)
ax.set_xticks(xtick)
xlable = []
for i in xtick:
xlable.append(str(rate.index[i]))
ax.set_xticklabels(xlable, rotation=90)
ext = math.floor(0.01 * len(rate))
ax.set_xlim((0 - ext, len(rate) + ext))
ax.plot(rate, color='blue', label='origin')
ax.legend(loc='best')
ax.set_title("M Rate Series")
# 计算自相关系数
plot_acf(rate, use_vlines=True, lags=30)
plot_acf(abs(rate), use_vlines=True, lags=30)
# 自相关
plot_acf(rate**2, use_vlines=True, lags=30)
# 偏相关
plot_pacf(rate, use_vlines=True, lags=30)
acfs = stattools.acf(rate)
# 计算偏自相关系数
pacfs = stattools.pacf(rate)
"""
1 通过建立均值方程或计量经济学模型，消除线性依赖
2 对方程残差进行ARCH效应检验
3 如果具有ARCH效应，建立波动率模型
4 检验拟合的模型，如有必要进行改进
"""
# 平稳性 1 看时序图 2 看自相关和偏自相关 3 单位根检验DF ADF PP检验
# ADF检验
adfrate = ADF(rate)
print(adfrate.summary().as_text())
plot_pacf(rate**2, use_vlines=True, lags=30)
model3 = arch.arch_model(rate, mean='Constant', vol='ARCH',p=3)
res = model3.fit()
res.plot()
model3 = arch.arch_model(rate, mean='Constant', vol='ARCH', p=10)
res = model3.fit()
res.plot()
model3 = arch.arch_model(rate, mean='Constant', vol='GARCH')
res = model3.fit()
res.plot()
close = self.data['收盘价']
fig = plt.figure(figsize=(20, 10), dpi=80, facecolor=[199 / 255, 238 / 255, 206 / 255], edgecolor='g')
fig.subplots_adjust(left=0.05, right=0.95, top=0.95, bottom=0.15)
ax = fig.add_subplot(1, 1, 1)
ax.set_facecolor([199 / 255, 238 / 255, 206 / 255])
min_ticks = math.floor(len(close) / 30)
ax.set_xlabel('time')
ax.set_ylabel('price')
xtick = []
for i in range(0, len(close), min_ticks):
xtick.append(i)
ax.set_xticks(xtick)
xlable = []
for i in xtick:
xlable.append(str(close.index[i]))
ax.set_xticklabels(xlable, rotation=90)
ext = math.floor(0.01 * len(close))
ax.set_xlim((0 - ext, len(close) + ext))
ax.plot(close, color='blue', label='close')
ax2 = ax.twinx()
pl2 = ax2.plot(res.conditional_volatility, color='m', linestyle='-', label='volatility')
ax.legend(loc=1)
ax2.legend(loc=2)
# 画收盘线，然后画收益率小概率尾部事件发生的位置
rate = res.conditional_volatility
rate_right = rate[rate > rate.quantile(0.95)]
for i in range(0, len(rate_right)):
x = rate[rate.index == rate_right.index[i]]
x[0] = ax.get_ylim()[0] * 1.05
ax.plot(x, color='red', marker='*')
x[0] = close[x.index[0]]
ax.plot(x, color='red', marker='*')
rate_left = rate[rate < rate.quantile(0.05)]
for i in range(0, len(rate_left)):
x = rate[rate.index == rate_left.index[i]]
x[0] = ax.get_ylim()[0] * 1.03
ax.plot(x, color='green', marker='^')
x[0] = close[x.index[0]]
ax.plot(x, color='green', marker='^')
ax.set_xticks(xtick)
ax.set_xticklabels(xlable, rotation=90)
ax.set_title("GARCH model tail vs close price")
