def select_Time_TRIX(self):
#EMA
ema_close_short = self.df_close[self.COL_EMA_S].get_values()
ema_ema_close_short = pd.ewma(ema_close_short, span=self.AVR_SHORT)
tr_close = pd.ewma(ema_ema_close_short, span=self.AVR_SHORT)
# ma_list = [self.AVR_SHORT, self.AVR_SHORT] #N,M
#
# if ma_list[0] == self.AVR_SHORT:
# ema_close = self.ema_short
# else:
# ema_close = pd.ewma(self.close_price, span=ma_list[0])
# ema_close = pd.ewma(ema_close, span=ma_list[0])
# tr_close = pd.ewma(ema_close, span=ma_list[0])
trixsList = [0]
for i in range(1, len(tr_close)):
#print tr_close[i], tr_close[i-1]
trix = (tr_close[i]-tr_close[i-1])/tr_close[i-1]*100
trixsList.append(trix)
trixs = np.array(trixsList)
maxtrix = pd.rolling_mean(trixs, self.AVR_LONG)
signal = SIGNAL_DEFAULT
if trixs[-1] > trixs[-2] and trixs[-1] > maxtrix[-1] \
and trixs[-2] < maxtrix[-2]:
signal = SIGNAL_BUY
elif trixs[-1] < trixs[-2] and trixs[-1] < maxtrix[-1] \
and trixs[-2] > maxtrix[-2]:
signal = SIGNAL_SALE
return signal
def storage(code):
try:
a=ts.get_hist_data(code)
except:
print('{} is wrong'.format(code))
else:
if a is not None:
a['code']=str(code)
a.sort_index(inplace=True)
boll(a)
a['rsi']=rsi(a)
kdj(a,9,3,3)
a['macd']=pd.ewma(a.close,12)-pd.ewma(a.close,26)
a['ma30']=pd.rolling_mean(a.close,30)
a['ma60']=pd.rolling_mean(a.close,60)
a['ma90']=pd.rolling_mean(a.close,90)
a['change30']=pd.rolling_std(np.gradient(a.ma30),30)
for t in [5,10,20,30,60,90]:
a['max'+str(t)]=pd.rolling_max(a.close,t)
a['min'+str(t)]=pd.rolling_min(a.close,t)
a['macd_a']=pd.ewma(a.close,12)
a['macd_d']=pd.ewma(a.close,26)
a['diff5']=100*(a.shift(-5).close-a.close)/a.close
a['diff10']=100*(a.shift(-10).close-a.close)/a.close
a['diff5_c']=a.diff5.apply(category)
a['diff10_c']=a.diff10.apply(category)
a.dropna()
return a
def calc_ewmac_forecast(price, Lfast, Lslow=None):
"""
Calculate the ewmac trading fule forecast, given a price and EWMA speeds Lfast, Lslow and vol_lookback
"""
# price: This is the stitched price series
# We can't use the price of the contract we're trading, or the volatility will be jumpy
# And we'll miss out on the rolldown. See
# http://qoppac.blogspot.co.uk/2015/05/systems-building-futures-rolling.html
price=price.resample("1B", how="last")
if Lslow is None:
Lslow = 4 * Lfast
# We don't need to calculate the decay parameter, just use the span
# directly
fast_ewma = pd.ewma(price, span=Lfast)
slow_ewma = pd.ewma(price, span=Lslow)
raw_ewmac = fast_ewma - slow_ewma
vol = robust_vol_calc(price.diff())
return divide_df_single_column(raw_ewmac, vol)
def fill_df_MACD_normalratio(df=None,params=None,names=['macd','signal','htg']):
if df is None: return None
if len(df.index)<40:
#print("index length=%d <40, return None"%len(index))
return None
#df=df.sort_index(ascending=True)
if params is None: params=MACD_STD_PARAMS
pos_scale=1
neg_scale=1
close=df.close
ema1=pd.ewma(close,span=params[0])
ema2=pd.ewma(close,span=params[1])
macd=(ema1-ema2)/ema2*100
signal=pd.ewma(macd,span=params[2])
htg=macd-signal
macd_max=macd.max()
macd_min=macd.min()
pos_scale=abs(1/macd_max)
neg_scale=abs(1/macd_min)
macd=macd.map(lambda x :x*neg_scale if x<0 else x*pos_scale)
signal=signal.map(lambda x :x*neg_scale if x<0 else x*pos_scale)
htg=htg.map(lambda x :x*neg_scale if x<0 else x*pos_scale)
#df['ema'+str(params[0])]=ema1
#df['ema'+str(params[1])]=ema2
df[names[0]]=macd
df[names[1]]=signal
df[names[2]]=htg
return df
def get_MACD_ratio(se,params=[12,26,9]):
ema1=pd.ewma(se,span=params[0])
ema2=pd.ewma(se,span=params[1])
macd=(ema1-ema2)/ema2*100
signal=pd.ewma(macd,span=params[2])
htg=macd-signal
return macd,signal,htg
def get_MACD(se,params=[12,26,9]):
ema1=pd.ewma(se,span=params[0])
ema2=pd.ewma(se,span=params[1])
macd=ema1-ema2
signal=pd.ewma(macd,span=params[2])
htg=macd-signal
return macd,signal,htg
def ADX(df, n, n_ADX):
i = 0
UpI = []
DoI = []
while i + 1 <= df.index[-1]:
UpMove = df.get_value(i + 1, 'High') - df.get_value(i, 'High')
DoMove = df.get_value(i, 'Low') - df.get_value(i + 1, 'Low')
if UpMove > DoMove and UpMove > 0:
UpD = UpMove
else: UpD = 0
UpI.append(UpD)
if DoMove > UpMove and DoMove > 0:
DoD = DoMove
else: DoD = 0
DoI.append(DoD)
i = i + 1
i = 0
TR_l = [0]
while i < df.index[-1]:
TR = max(df.get_value(i + 1, 'High'), df.get_value(i, 'Close')) - min(df.get_value(i + 1, 'Low'), df.get_value(i, 'Close'))
TR_l.append(TR)
i = i + 1
TR_s = pd.Series(TR_l)
ATR = pd.Series(pd.ewma(TR_s, span = n, min_periods = n))
UpI = pd.Series(UpI)
DoI = pd.Series(DoI)
PosDI = pd.Series(pd.ewma(UpI, span = n, min_periods = n - 1) / ATR)
NegDI = pd.Series(pd.ewma(DoI, span = n, min_periods = n - 1) / ATR)
ADX = pd.Series(pd.ewma(abs(PosDI - NegDI) / (PosDI + NegDI), span = n_ADX, min_periods = n_ADX - 1), name = 'ADX_' + str(n) + '_' + str(n_ADX))
df = df.join(ADX)
return df
def plot(name):
name = str(name)
data_ext = YFHistDataExtr()
data_ext.set_interval_to_retrieve(200)
data_ext.set_multiple_stock_list([name])
data_ext.get_hist_data_of_all_target_stocks()
# convert the date column to date object
data_ext.all_stock_df['Date'] = pandas.to_datetime( data_ext.all_stock_df['Date'])
temp_data_set = data_ext.all_stock_df.sort('Date', ascending=True)
temp_data_set['20d_ma'] = pandas.rolling_mean(temp_data_set['Adj Close'], window=20)
temp_data_set['50d_ma'] = pandas.rolling_mean(temp_data_set['Adj Close'], window=50)
temp_data_set['Bol_upper'] = pandas.rolling_mean(temp_data_set['Adj Close'], window=20) + 2* pandas.rolling_std(temp_data_set['Adj Close'], 20, min_periods=20)
temp_data_set['Bol_lower'] = pandas.rolling_mean(temp_data_set['Adj Close'], window=20) - 2* pandas.rolling_std(temp_data_set['Adj Close'], 20, min_periods=20)
temp_data_set['Bol_BW'] = ((temp_data_set['Bol_upper'] - temp_data_set['Bol_lower'])/temp_data_set['20d_ma'])*100
temp_data_set['Bol_BW_200MA'] = pandas.rolling_mean(temp_data_set['Bol_BW'], window=50)
temp_data_set['Bol_BW_200MA'] = temp_data_set['Bol_BW_200MA'].fillna(method='backfill')
temp_data_set['20d_exma'] = pandas.ewma(temp_data_set['Adj Close'], span=20)
temp_data_set['50d_exma'] = pandas.ewma(temp_data_set['Adj Close'], span=50)
data_ext.all_stock_df = temp_data_set.sort('Date', ascending = False)
data_ext.all_stock_df.plot(x='Date', y=['Adj Close', '20d_ma', '50d_ma', 'Bol_upper', 'Bol_lower'])
cur_path = os.path.dirname(os.path.abspath(__file__))
#cur_path = os.path.join(cur_path, "app/analysis")
data_path = os.path.join(cur_path, "raw_stock_data")
print cur_path
img_path = os.path.join(cur_path, "static/img")
history_img_path = os.path.join(img_path, "history.png")
dividend_img_path = os.path.join(img_path, "dividend.png")
plt.savefig(history_img_path)
data_ext.all_stock_df.plot(x='Date', y=['Bol_BW','Bol_BW_200MA'])
plt.savefig(dividend_img_path)
return render_template('analysis.html')
def RSI(df, n):
"""
Relative Strength Index
"""
i = 0
UpI = [0]
DoI = [0]
while i + 1 <= len(df) - 1: # df.index[-1]
UpMove = df.get_value(i + 1, 'High') - df.get_value(i, 'High')
DoMove = df.get_value(i, 'Low') - df.get_value(i + 1, 'Low')
if UpMove > DoMove and UpMove > 0:
UpD = UpMove
else: UpD = 0
UpI.append(UpD)
if DoMove > UpMove and DoMove > 0:
DoD = DoMove
else: DoD = 0
DoI.append(DoD)
i = i + 1
UpI = pd.Series(UpI)
DoI = pd.Series(DoI)
PosDI = pd.Series(pd.ewma(UpI, span = n, min_periods = n - 1))
NegDI = pd.Series(pd.ewma(DoI, span = n, min_periods = n - 1))
result = pd.Series(PosDI / (PosDI + NegDI), name = 'RSI_' + str(n))
return out(SETTINGS, df, result)
def Chaikin(df):
"""
Chaikin Oscillator
"""
ad = (2 * df['Close'] - df['High'] - df['Low']) / (df['High'] - df['Low']) * df['Volume']
result = pd.Series(pd.ewma(ad, span = 3, min_periods = 2) - pd.ewma(ad, span = 10, min_periods = 9), name = 'Chaikin')
return out(SETTINGS, df, result)
def TSI(df, r, s,ksgn='close'):
'''
def TSI(df, r, s,ksgn='close'):
TSI，真实强度指数,True Strength Index
TSI是相对强弱指数 (RSI) 的变体。
TSI 使用价格动量的双重平滑指数移动平均线，剔除价格的震荡变化并发现趋势的变化。
r一般取25，是一般取13
【输入】
df, pd.dataframe格式数据源
r,s，时间长度; r一般取25，是一般取13
ksgn，列名，一般是：close收盘价
【输出】
df, pd.dataframe格式数据源,
增加了一栏：tsi，输出数据
'''
xnam='tsi'.format(d=r,d2=s)
M = pd.Series(df[ksgn].diff(1))
aM = abs(M)
EMA1 = pd.Series(pd.ewma(M, span = r, min_periods = r - 1))
aEMA1 = pd.Series(pd.ewma(aM, span = r, min_periods = r - 1))
EMA2 = pd.Series(pd.ewma(EMA1, span = s, min_periods = s - 1))
aEMA2 = pd.Series(pd.ewma(aEMA1, span = s, min_periods = s - 1))
TSI = pd.Series(EMA2 / aEMA2, name = xnam)
df = df.join(TSI)
return df
def ADX(df, n, n_ADX):
"""
Average Directional Movement Index
"""
i = 0
UpI = []
DoI = []
while i + 1 <= len(df) - 1: #df.index[-1]:
UpMove = df.get_value(i + 1, 'High') - df.get_value(i, 'High')
DoMove = df.get_value(i, 'Low') - df.get_value(i + 1, 'Low')
if UpMove > DoMove and UpMove > 0:
UpD = UpMove
else: UpD = 0
UpI.append(UpD)
if DoMove > UpMove and DoMove > 0:
DoD = DoMove
else: DoD = 0
DoI.append(DoD)
i = i + 1
i = 0
TR_l = [0]
while i < len(df) - 1: #df.index[-1]:
TR = max(df.get_value(i + 1, 'High'), df.get_value(i, 'Close')) - min(df.get_value(i + 1, 'Low'), df.get_value(i, 'Close'))
TR_l.append(TR)
i = i + 1
TR_s = pd.Series(TR_l)
ATR = pd.Series(pd.ewma(TR_s, span = n, min_periods = n))
UpI = pd.Series(UpI)
DoI = pd.Series(DoI)
PosDI = pd.Series(pd.ewma(UpI, span = n, min_periods = n - 1) / ATR)
NegDI = pd.Series(pd.ewma(DoI, span = n, min_periods = n - 1) / ATR)
result = pd.Series(pd.ewma(abs(PosDI - NegDI) / (PosDI + NegDI), span = n_ADX, min_periods = n_ADX - 1), name = 'ADX_' + str(n) + '_' + str(n_ADX))
return out(SETTINGS, df, result)
def MACD(df, n_fast, n_slow,ksgn='close'):
'''
def MACD(df, n_fast, n_slow):
#MACD指标信号和MACD的区别, MACD Signal and MACD difference
MACD是查拉尔·阿佩尔(Geral Appel)于1979年提出的，由一快及一慢指数移动平均（EMA）之间的差计算出来。
“快”指短时期的EMA，而“慢”则指长时期的EMA，最常用的是12及26日EMA：
【输入】
df, pd.dataframe格式数据源
n，时间长度
ksgn，列名，一般是：close收盘价
【输出】
df, pd.dataframe格式数据源,
增加了3栏：macd,sign,mdiff
'''
xnam='macd'.format(n=n_fast,n2=n_slow)
xnam2='msign'.format(n=n_fast,n2=n_slow)
xnam3='mdiff'.format(n=n_fast,n2=n_slow)
EMAfast = pd.Series(pd.ewma(df[ksgn], span = n_fast, min_periods = n_slow - 1))
EMAslow = pd.Series(pd.ewma(df[ksgn], span = n_slow, min_periods = n_slow - 1))
MACD = pd.Series(EMAfast - EMAslow, name = xnam)
MACDsign = pd.Series(pd.ewma(MACD, span = 9, min_periods = 8), name =xnam2)
MACDdiff = pd.Series(MACD - MACDsign, name =xnam3)
df = df.join(MACD)
df = df.join(MACDsign)
df = df.join(MACDdiff)
return df
def TRIX(df, n,ksgn='close'):
'''
def TRIX(df, n,ksgn='close'):
【输入】
df, pd.dataframe格式数据源
n，时间长度
ksgn，列名，一般是：close收盘价
【输出】
df, pd.dataframe格式数据源,
增加了一栏：trix_{n}，输出数据
'''
xnam='trix_{n}'.format(n=n)
EX1 = pd.ewma(df[ksgn], span=n, min_periods=n - 1)
EX2 = pd.ewma(EX1, span=n, min_periods=n - 1)
EX3 = pd.ewma(EX2, span=n, min_periods=n - 1)
i = 0
ROC_l = [0]
while i + 1 <= len(df) - 1: # df.index[-1]:
ROC = (EX3[i + 1] - EX3[i]) / EX3[i]
ROC_l.append(ROC)
i = i + 1
trix = pd.Series(ROC_l, name=xnam)
#df = df.join(trix)
trix.index=df.index;
df[xnam]=trix
#print(trix.tail())
#print('n',len(df))
return df
def MassI(df):
Range = df['high'] - df['low']
EX1 = pd.ewma(Range, span = 9, min_periods = 8)
EX2 = pd.ewma(EX1, span = 9, min_periods = 8)
Mass = EX1 / EX2
MassI = pd.Series(pd.rolling_sum(Mass, 25), name = 'MassIndex')
return MassI
def TEMA(ts, n):
n = int(n)
ts_ema1 = pd.Series( pd.ewma(ts, span = n, adjust = False), name = 'EMA_' + str(n) )
ts_ema2 = pd.Series( pd.ewma(ts_ema1, span = n, adjust = False), name = 'EMA2_' + str(n) )
ts_ema3 = pd.Series( pd.ewma(ts_ema2, span = n, adjust = False), name = 'EMA3_' + str(n) )
ts_tema = pd.Series( 3 * ts_ema1 - 3 * ts_ema2 + ts_ema3, name = 'TEMA_' + str(n) )
return ts_tema
def indicators(prices, window=20):
# Calculate out the Moving Average and the SD of the prices
sma = pd.stats.moments.rolling_mean(prices, window)
sd = pd.stats.moments.rolling_std(prices, window)
# Ratio to the bolligner bands (above upper band is > 1; below lower band <-1)
bb_diff = (prices - sma)/(2*sd)
# Ratio to a recent high in a specific window
winmax = pd.stats.moments.rolling_max(prices, window)
high_diff = (prices-winmax)/(2*sd)
# Ratio to a recent low in a specific window
winlow = pd.stats.moments.rolling_min(prices, window)
low_diff = (prices-winlow)/(2*sd)
#Moving Average Convergence Divergence (MACD) using exponentially weighted Moving average
macd=pd.ewma(prices,span=window) - pd.ewma(prices,window/2)
# combine all to a dataframe
df = pd.DataFrame(pd.concat([bb_diff, high_diff, low_diff, macd], axis=1))
df.columns = ['bb_diff', 'high_diff', 'low_diff', 'macd']
df.columns = [prices.columns.values[0] + '_' + x for x in df.columns]
# Normalaise the data to between -1 and 1
df = df.apply(lambda x: ((x - np.min(x)))*2 / (np.max(x) - np.min(x))-1)
return df[[0,1,2,3]]
def MACD(df, n_fast, n_slow):
EMAfast = pd.Series(pd.ewma(df['close'], span = n_fast, min_periods = n_slow - 1))
EMAslow = pd.Series(pd.ewma(df['close'], span = n_slow, min_periods = n_slow - 1))
MACD = pd.Series(EMAfast - EMAslow, name = 'MACD_' + str(n_fast) + '_' + str(n_slow))
MACDsign = pd.Series(pd.ewma(MACD, span = 9, min_periods = 8), name = 'MACDsign_' + str(n_fast) + '_' + str(n_slow))
MACDdiff = pd.Series(MACD - MACDsign, name = 'MACDdiff_' + str(n_fast) + '_' + str(n_slow))
return pd.concat([MACD, MACDsign, MACDdiff], join='outer', axis=1)
def processEMA(stockCsvPath, stockCsvNewPath):
#导入数据，stockCsvPath为在电脑中的路径
stock_data = pd.read_csv(stockCsvPath)
# 将数据按照交易日期从远到近排序
stock_data.sort('Date', inplace=True)
#=====================计算移动平均线
# 分别计算5日、20日、60日移动平均线
ma_list = [5, 20, 60]
# 计算简单算术移动平均线MA - 注意：stock_data['close']为股票每条的收盘价
for ma in ma_list:
stock_data['MA_' + str(ma)] = pd.rolling_mean(stock_data['Adj Close'], ma)
# 计算指数平滑移动平均线EMA
for ma in ma_list:
stock_data['EMA_' + str(ma)] = pd.ewma(stock_data['Adj Close'], span=ma)
# 将数据按照交易日期从近到远排序
stock_data.sort('Date', ascending=False, inplace=True)
stock_data['DIF'] = stock_data['EMA_'+str(ma_list[0])] - stock_data['EMA_'+str(ma_list[-1])]
stock_data['DEA_' + str(10)] = pd.ewma(stock_data['DIF'], span=10)
# =================================== 将算好的数据输出到csv文件，这里请填写输出文件在您电脑的路径
stock_data.to_csv(stockCsvNewPath, index=False)
def select_Time_TRIX(stock_data, stockName):
start_money = 100000000
now_count = 0
now_money = start_money
# 将数据按照交易日期从远到近排序
stock_data.sort('Date', inplace=True)
close_price = stock_data['Adj Close'].get_values()
#EMA
ma_list = [AVR_SHORT, AVR_SHORT] #N,M
ema_close = pd.ewma(close_price, span=ma_list[0])
ema_close = pd.ewma(ema_close, span=ma_list[0])
tr_close = pd.ewma(ema_close, span=ma_list[0])
trixsList = [0]
for i in range(1, len(tr_close)):
#print tr_close[i], tr_close[i-1]
trix = (tr_close[i]-tr_close[i-1])/tr_close[i-1]*100
trixsList.append(trix)
trixs = np.array(trixsList)
maxtrix = pd.rolling_mean(trixs, ma_list[1])
signals = [0]*(ma_list[1]+1)
tradeTimes = 0
bBuySignal = True
for t in range(ma_list[1]+1, len(close_price)):
signal = SIGNAL_DEFAULT
if trixs[t] > trixs[t-1] and trixs[t] > maxtrix[t] \
and trixs[t-1] < maxtrix[t-1]:
if bBuySignal:
signal = SIGNAL_BUY
now_count = (int)(now_money / close_price[t] /100)*100
now_money = start_money - now_count * close_price[t]
#print u"股票价格/持股数/剩余金额：",close_price[t], '/', now_count, '/', now_money
bBuySignal = False
elif trixs[t] < trixs[t-1] and trixs[t] < maxtrix[t] \
and trixs[t-1] > maxtrix[t-1]:
if bBuySignal == False:
signal = SIGNAL_SALE
now_money += now_count * close_price[t]
now_count = 0
#print u"股票价格/持股数/剩余金额：",close_price[t], '/', now_count, '/', now_money
bBuySignal = True
signals.append(signal)
if signal != 0:
#print 't:', t, ' signal:', signal
tradeTimes += 1
print stockName, u"收益率：", (now_money+now_count * close_price[-1]-start_money)/start_money*100, '%\t' \
u"交易次数", tradeTimes, u" 最新市值：", now_money+now_count * close_price[-1]
stock_data['SIGNAL_TRIX'] = signals
def select_Time_MACD(self):
#EMA
ma_list = [self.AVR_SHORT, self.AVR_LONG]
ma_dea = 10
if ma_list[0] == self.AVR_SHORT and ma_list[1] == self.AVR_LONG:
ema_close_short = self.ema_12
ema_close_long = self.ema_40
else:
ema_close_short = pd.ewma(self.close_price, span=ma_list[0])
ema_close_long = pd.ewma(self.close_price, span=ma_list[1])
dif_price = ema_close_short - ema_close_long
dea_price = pd.ewma(dif_price, span=ma_dea)
macd_price = 2 * (dif_price - dea_price)
signal = SIGNAL_DEFAULT
if dif_price[-1] > dif_price[-2] and dif_price[-1] > dea_price[-2] \
and dif_price[-2] < dea_price[-2] and dea_price[-1] > 0:
signal = SIGNAL_BUY
elif dif_price[-1] < dif_price[-2] and dif_price[-1] < dea_price[-1] \
and dif_price[-2] > dea_price[-2] and dif_price[-1] < 0:
signal = SIGNAL_SALE
return signal
def select_Time_TRIX(self):
#EMA
ma_list = [self.AVR_SHORT, self.AVR_SHORT] #N,M
if ma_list[0] == self.AVR_SHORT:
ema_close = self.ema_12
else:
ema_close = pd.ewma(self.close_price, span=ma_list[0])
ema_close = pd.ewma(ema_close, span=ma_list[0])
tr_close = pd.ewma(ema_close, span=ma_list[0])
trixsList = [0]
for i in range(1, len(tr_close)):
#print tr_close[i], tr_close[i-1]
trix = (tr_close[i]-tr_close[i-1])/tr_close[i-1]*100
trixsList.append(trix)
trixs = np.array(trixsList)
maxtrix = pd.rolling_mean(trixs, ma_list[1])
signal = SIGNAL_DEFAULT
if trixs[-1] > trixs[-2] and trixs[-1] > maxtrix[-1] \
and trixs[-2] < maxtrix[-2]:
signal = SIGNAL_BUY
elif trixs[-1] < trixs[-2] and trixs[-1] < maxtrix[-1] \
and trixs[-2] > maxtrix[-2]:
signal = SIGNAL_SALE
return signal
def MACD( ohlc, f=12, s=26, m=9):
fast_ema = pd.ewma( ohlc.close, span=f)
slow_ema = pd.ewma( ohlc.close, span=s)
macd = fast_ema - slow_ema
macd_sig = pd.ewma( macd, span=9)
hist = macd - macd_sig
return pd.DataFrame( {"MACD_hist":hist, "MACD":macd_sig}, index=[ohlc.index])
def ELI( ohlc, n=14):
""" This is the Ehler's Leading Indicator ... it predicts cyclical changes
in price using a tripple EMA scheme. It first calculates two EMAs on the
original price: one that's 1/2 n and the other 1/4 n. Then those are
subtracted, giving us a synthetic price. This is smoothed with an EMA
of 1/4 n. The ELI is calculated by subtracting the syn from the syn's EMA.
This leads cyclical changes, assuming they exist in the data.
Parameters:
ohlc : OHLC dataframe to take our prices from. Right now we're using close,
but theoretically this could be anything.
n : n-period cycles in our data. Tuning this parameter totally depends
on the market/dataset, but I found 14 to work well for now.
"""
a = n / 4.0
_EMA1 = pd.ewma( ohlc.close, span=a)
a = n / 2.0
_EMA2 = pd.ewma( ohlc.close, span=a)
syn = _EMA1 - _EMA2
_EMAsyn = pd.ewma( syn, span=a)
_ELI = syn - _EMAsyn
return pd.DataFrame( { "ELI_%s"%n: _ELI }, index=[ohlc.index])
def RSI(df, n):
i = 0
UpI = [0]
DoI = [0]
while i + 1 <= df.index[-1]:
UpMove = df.get_value(i + 1, "High") - df.get_value(i, "High")
DoMove = df.get_value(i, "Low") - df.get_value(i + 1, "Low")
if UpMove > DoMove and UpMove > 0:
UpD = UpMove
else:
UpD = 0
UpI.append(UpD)
if DoMove > UpMove and DoMove > 0:
DoD = DoMove
else:
DoD = 0
DoI.append(DoD)
i = i + 1
UpI = Series(UpI)
DoI = Series(DoI)
PosDI = Series(ewma(UpI, span=n, min_periods=n - 1))
NegDI = Series(ewma(DoI, span=n, min_periods=n - 1))
RSI = Series(PosDI / (PosDI + NegDI), name="RSI_" + str(n))
df = df.join(RSI)
return df
def computeBidAskVolumeFeats(df, feats_df):
df['askbid_vol_ema_200ms'] = df['ask_vol_ema_200ms'] - df['bid_vol_ema_200ms']
df['askbid_vol_ema_2s'] = df['ask_vol_ema_2s'] - df['bid_vol_ema_2s']
df['askbid_vol_ema_10s'] = df['ask_vol_ema_10s'] - df['bid_vol_ema_10s']
feats_df['diff_200ms_2s_askbid_vol_ema'] = df['askbid_vol_ema_200ms'] - df['askbid_vol_ema_2s'] #1
feats_df['diff_2s_10s_askbid_vol_ema'] = df['askbid_vol_ema_2s'] - df['askbid_vol_ema_10s'] #1
feats_df['d_diff_200ms_2s_askbid_vol_ema'] = feats_df['diff_200ms_2s_askbid_vol_ema'] - feats_df['diff_200ms_2s_askbid_vol_ema'].shift(1) #1
feats_df['d_diff_2s_10s_askbid_vol_ema'] = feats_df['diff_2s_10s_askbid_vol_ema'] - feats_df['diff_2s_10s_askbid_vol_ema'].shift(2) #1
feats_df['norm_askbid_vol_ema_200ms'] = (df['ask_vol_ema_200ms'] - df['bid_vol_ema_200ms']) / (df['ask_vol_ema_200ms'] + df['bid_vol_ema_200ms']) #1
feats_df['norm_askbid_vol_ema_2s'] = (df['ask_vol_ema_2s'] - df['bid_vol_ema_2s']) / (df['ask_vol_ema_2s'] + df['bid_vol_ema_2s']) #1
feats_df['norm_askbid_vol_ema_10s'] = (df['ask_vol_ema_10s'] - df['bid_vol_ema_10s']) / (df['ask_vol_ema_10s'] + df['bid_vol_ema_10s']) #1
feats_df['diff_norm_200ms_2s_askbid_vol_ema'] = feats_df['norm_askbid_vol_ema_200ms'] - feats_df['norm_askbid_vol_ema_2s'] #1
feats_df['diff_norm_2s_10s_askbid_vol_ema'] = feats_df['norm_askbid_vol_ema_2s'] - feats_df['norm_askbid_vol_ema_10s'] #1
df['smooth_norm_askbid_vol_ema_200ms'] = pd.ewma(feats_df['norm_askbid_vol_ema_200ms'], span=25)
df['smooth_norm_askbid_vol_ema_2s'] = pd.ewma(feats_df['norm_askbid_vol_ema_2s'], span=25)
df['smooth_norm_askbid_vol_ema_10s'] = pd.ewma(feats_df['norm_askbid_vol_ema_10s'], span=25)
feats_df['smooth_diff_norm_200ms_2s_askbid_vol_ema'] = pd.ewma(feats_df['diff_norm_200ms_2s_askbid_vol_ema'], span=25) #1
feats_df['smooth_diff_norm_2s_10s_askbid_vol_ema'] = pd.ewma(feats_df['diff_norm_2s_10s_askbid_vol_ema'], span=25) #1
feats_df['d_smooth_diff_norm_200ms_2s_askbid_vol_ema'] = feats_df['smooth_diff_norm_200ms_2s_askbid_vol_ema'] - feats_df['smooth_diff_norm_200ms_2s_askbid_vol_ema'].shift(1) #1
feats_df['d_smooth_diff_norm_2s_10s_askbid_vol_ema'] = feats_df['smooth_diff_norm_2s_10s_askbid_vol_ema'] - feats_df['smooth_diff_norm_2s_10s_askbid_vol_ema'].shift(5) #1
return feats_df
def MACD(df, n_fast, n_slow, price='Close'):
"""
MACD, MACD Signal and MACD difference
"""
EMAfast = pd.Series(
pd.ewma(
df[price],
span=n_fast,
min_periods=n_slow - 1
)
)
EMAslow = pd.Series(
pd.ewma(
df[price],
span=n_slow,
min_periods=n_slow - 1
)
)
MACD = pd.Series(EMAfast - EMAslow, name='MACD_%d_%d' % (n_fast, n_slow))
MACDsign = pd.Series(
pd.ewma(
MACD,
span=9,
min_periods=8
),
name='MACDsign_%d_%d' % (n_fast, n_slow)
)
MACDdiff = pd.Series(
MACD - MACDsign,
name='MACDdiff_%d_%d' % (n_fast, n_slow)
)
result = pd.DataFrame([MACD, MACDsign, MACDdiff]).transpose()
return out(SETTINGS, df, result)
def get_signals(data, long_window, short_window, confirmation_window):
returns = data['alpha'].cumsum()
_short = pd.ewma(returns, span=short_window)
_long = pd.ewma(returns, span=long_window)
result = _short > _long
result = _adjust_series_of_signals(result, confirmation_window)
return remove_consecutive_values(result)
def calc_ewmac_forecast(price, Lfast, Lslow=None, usescalar=True):
"""
Calculate the ewmac trading fule forecast, given a price and EWMA speeds Lfast, Lslow and vol_lookback
Assumes that 'price' is daily data
"""
## price: This is the stitched price series
## We can't use the price of the contract we're trading, or the volatility will be jumpy
## And we'll miss out on the rolldown. See http://qoppac.blogspot.co.uk/2015/05/systems-building-futures-rolling.html
if Lslow is None:
Lslow=4*Lfast
## We don't need to calculate the decay parameter, just use the span directly
fast_ewma=pd.ewma(price, span=Lfast)
slow_ewma=pd.ewma(price, span=Lslow)
raw_ewmac=fast_ewma - slow_ewma
## volatility adjustment
stdev_returns=volatility(price)
vol_adj_ewmac=raw_ewmac/stdev_returns
## scaling adjustment
if usescalar:
f_scalar=ewmac_forecast_scalar(Lfast, Lslow)
forecast=vol_adj_ewmac*f_scalar
else:
forecast=vol_adj_ewmac
cap_forecast=cap_series(forecast, capmin=-20.0,capmax=20.0)
return cap_forecast
def calc_ewmac_forecast(price, Lfast, Lslow):
price=price.resample("1B", how="last")
fast_ewma = pd.ewma(price, span=Lfast)
slow_ewma = pd.ewma(price, span=Lslow)
raw_ewmac = fast_ewma - slow_ewma
vol = robust_vol_calc(price.diff())
return raw_ewmac / vol
def plot_agg_comparison(LD,
value,
groupby,
name_legend_map,
sm=5,
n_quantiles=10):
'''Make Bokeh time series plot.
INPUTS:
LD: pandas dataframe with loan data
value: name of column to plot as response variable.
groupby: name of column to use for grouping data.
sm: number of months to use for EWMA smoothing window (default 5).
n_quantiles: number of quantiles to use if plotting response variable by quantiles.
OUTPUTS:
(script, div) html components of Bokeh plot.'''
min_counts = 500 #min number of loans for each group (across all time points) for inclusion in plot
if value == 'counts': #treat response variable 'counts' by setting that to the agg fnx.
agg = 'count'
value = 'ROI'
else:
agg = 'mean'
if groupby == 'loan_status': #specify order and pallete if plotting by loan-status
group_set = [
'Fully Paid', 'Current', 'In Grace Period', 'Late (16-30 days)',
'Late (31-120 days)', 'Default', 'Charged Off'
]
pal = sns.cubehelix_palette(n_colors=len(group_set)).as_hex()
elif groupby == 'quantiles': #same for if plotting by quantiles
quantiles = np.linspace(n_quantiles, 100 - n_quantiles,
n_quantiles).astype(float) / 100.
pal = sns.cubehelix_palette(n_colors=len(quantiles)).as_hex()
else: #otherwise, assume categorical grouping vars with no specified plotting order
group_set = np.sort(LD[groupby].unique())
pal = sns.color_palette("muted", n_colors=len(group_set)).as_hex()
start_date = np.datetime64(dt.datetime(
2009, 01, 01)) #only plot data for loans issued after this time
if groupby == 'quantiles':
time_avgs = LD[LD.issue_d > start_date].groupby(
['issue_d'])[value].quantile(quantiles)
time_avgs = time_avgs.swaplevel(0, 1) # swap quantile and time levels
group_set = time_avgs.index.levels[0]
else:
time_avgs = LD[LD.issue_d > start_date].groupby([groupby, 'issue_d'
])[value].agg(agg)
time_counts = LD[LD.issue_d > start_date].groupby([groupby, 'issue_d'
])[value].count()
time_avgs.index.names = [groupby, 'issue_d']
ylabel = name_legend_map[value]
if sm > 0: #apply EWMA smoothing to time-grouped avgs
time_avgs = pd.ewma(time_avgs, span=sm)
s1 = figure(x_axis_type="datetime",
plot_width=800,
plot_height=500,
tools=tools,
x_axis_label='Issue Date',
y_axis_label=ylabel)
for idx, group in enumerate(group_set):
if groupby != 'quantiles':
if time_counts.ix[group].sum() >= min_counts:
group_avg = time_avgs.ix[group]
s1.line(group_avg.index,
group_avg,
line_color=pal[idx],
line_width=4,
legend=str(group))
else:
group_avg = time_avgs.ix[group]
s1.line(group_avg.index,
group_avg,
line_color=pal[idx],
line_width=4,
legend=str(group))
s1.legend.label_text_font_size = '13'
s1.legend.orientation = "bottom_right"
return components(s1)
df = df.set_index('DATE')
PRICE0[prod] = df
PRICE = PRICE0
# calculate price series momentum (10-day), RSI (14-day), and MFI (14-days)
print('Calculate secondary series')
for prod in PRICE.keys():
previous = PRICE[prod].shift(periods=1)
# RSI
PRICE[prod]['UPMOVE'] = \
(PRICE[prod]['CLOSE'] > previous['CLOSE']) * \
(PRICE[prod]['CLOSE'] - previous['CLOSE'])
PRICE[prod]['DOWNMOVE'] = \
(PRICE[prod]['CLOSE'] < previous['CLOSE']) * \
(previous['CLOSE'] - PRICE[prod]['CLOSE'])
U = pd.ewma(PRICE[prod]['UPMOVE'], com=6.5)
D = pd.ewma(PRICE[prod]['DOWNMOVE'], com=6.5)
RS = U / D
PRICE[prod]['RSI'] = 100 - 100. / (1 + RS)
PRICE[prod].drop('UPMOVE', axis=1, inplace=True)
PRICE[prod].drop('DOWNMOVE', axis=1, inplace=True)
# MFI
PRICE[prod]['TP'] = (PRICE[prod]['HIGH'] + PRICE[prod]['LOW'] +
PRICE[prod]['CLOSE']) / 3.
PRICE[prod]['POSFLOW'] = \
(PRICE[prod]['CLOSE'] > previous['CLOSE']) * \
PRICE[prod]['TP'] * PRICE[prod]['VOLUME']
PRICE[prod]['NEGFLOW'] = \
(PRICE[prod]['CLOSE'] < previous['CLOSE']) * \
PRICE[prod]['TP'] * PRICE[prod]['VOLUME']
POS = pd.ewma(PRICE[prod]['POSFLOW'], com=6.5)
def ewma_metric(cls, n, df_price, name, adjust=True):
ewma_df = pd.ewma(df_price[[name]], n, adjust=adjust)
ewma_mt = list(ewma_df[name])
return ewma_mt
])
for key, value in dftest[4].items():
dfoutput['Critical value (%s)' % key] = value
print(dfoutput)
test_stationarity(data_label)
#处理时间序列：
#让时序数据变得稳定，去除趋势：平滑： 以一个滑动窗口内的均值代替原来的值，为了使值之间的差距缩小
#而在许多情况下，可以认为越近的时刻越重要。所以引入指数加权移动平均--
# Exponentially-weighted moving average.（pandas中通过ewma()函数提供了此功能。）
data_label_log = np.log(data_label)
# halflife的值决定了衰减因子alpha： alpha = 1 - exp(log(0.5) / halflife)
expmoving_avg = pd.ewma(data_label_log, 200)
data_label_log_moving_diff = data_label_log - expmoving_avg
print('aaaaaa')
print('data_label_log_moving_diff.type', data_label_log_moving_diff.dtype)
print('data_label_log_moving_diff:', data_label_log_moving_diff)
data_label_log_moving_diff.dropna(inplace=True)
test_stationarity(data_label_log_moving_diff)
#第四步 对时序数据进行预测
#一阶差分后数据已经稳定，所以d=1。所以用一阶差分化的data_label_log_moving_diff = data_label_log-expmoving_avg 作为输入
#先画出ACF,PACF的图像
from statsmodels.tsa.stattools import acf, pacf
#from statsmodels.graphics.tsaplots import acf, pacf
lag_acf = acf(data_label_log_moving_diff, nlags=20)
def ewma(df_in, periods):
if abs(periods) < 2:
return df_in
else:
return pd.ewma(df_in, abs(periods), min_periods=abs(periods))
for m in mkts.keys():
try:
data_index[m] = quandl.get(mkts[m], authtoken=token).Last
except:
try:
data_index[m] = quandl.get(mkts[m], authtoken=token).Settle
except:
try:
data_index[m] = quandl.get(mkts[m], authtoken=token).Value
except:
try:
data_index[m] = quandl.get(mkts[m], authtoken=token).value
except:
try:
data_index[m] = quandl.get(mkts[m],
authtoken=token).Rate
except:
print(m)
data_pct = data_index.pct_change()
mu = pd.ewma(data_pct, 60)
sd = pd.ewmstd(data_pct, 60)
zscores = (data_pct - mu) / sd
last = zscores.iloc[-2].dropna().sort_values()
last.plot(kind='barh', colormap='jet',
ylim=[-3, 3]).get_figure().savefig('zscore.png', bbox_inches='tight')
e = Email(to='*****@*****.**',
subject='Morning Update: Macro Dashboard')
e.add_attachment('zscore.png')
e.send()
def EMA(df, n):
EMA = pd.Series(pd.ewma(df['Close'], span=n, min_periods=n - 1),
name='EMA_' + str(n))
df = df.join(EMA)
return df
def feature_engineering(df, complete_dates):
df = df.groupby( ["Ciclo_Estacion", "day_counter", "ITERATION", "iteration_start", "iteration_end"])["hora"].count()
df = df.sort_index()
df = df.reset_index()
df = df.rename( columns = {"hora": "flow"})
df_append = pd.DataFrame()
for station in df.Ciclo_Estacion.unique():
df_station = df[df.Ciclo_Estacion == station]
df_merge = complete_dates.merge( df_station, on= ["day_counter", "ITERATION", "iteration_start", "iteration_end"], how ="left" )
df_merge["Ciclo_Estacion"] = station
df_merge.loc[pd.isnull(df_merge.flow), "flow"] =0
if len(df_append) ==0 :
df_append = df_merge
else:
df_append = df_append.append(df_merge)
df = df_append
#ITERATION (15 minutes) LAG VALUES
df["flow_lag1"] = df.groupby(["Ciclo_Estacion"])["flow"].shift(1)
df["flow_lag2"] = df.groupby(["Ciclo_Estacion"])["flow"].shift(2)
df["flow_lag3"] = df.groupby(["Ciclo_Estacion"])["flow"].shift(3)
df["flow_lag4"] = df.groupby(["Ciclo_Estacion"])["flow"].shift(4)
df["flow_lag5"] = df.groupby(["Ciclo_Estacion"])["flow"].shift(5)
df["flow_lag6"] = df.groupby(["Ciclo_Estacion"])["flow"].shift(6)
df["flow_lag7"] = df.groupby(["Ciclo_Estacion"])["flow"].shift(7)
df["flow_lag8"] = df.groupby(["Ciclo_Estacion"])["flow"].shift(8)
df["flow_rollingmean_lag1_4"] = pd.rolling_mean( df["flow_lag1"], 4)
df["flow_rollingmean_lag1_8"] = pd.rolling_mean( df["flow_lag1"], 8)
df["flow_rollingmean_lag1_12"] = pd.rolling_mean( df["flow_lag1"], 12)
df["flow_rollingmean_lag1_16"] = pd.rolling_mean( df["flow_lag1"], 16)
df["flow_rollingmean_lag4_4"] = pd.rolling_mean( df["flow_lag4"], 4)
df["flow_rollingmean_lag4_8"] = pd.rolling_mean( df["flow_lag4"], 8)
df["flow_rollingmean_lag4_12"] = pd.rolling_mean( df["flow_lag4"], 12)
df["flow_rollingmean_lag4_16"] = pd.rolling_mean( df["flow_lag4"], 16)
df["flow_rollingmean_lag8_4"] = pd.rolling_mean( df["flow_lag8"], 4)
df["flow_rollingmean_lag8_8"] = pd.rolling_mean( df["flow_lag8"], 8)
df["flow_rollingmean_lag8_12"] = pd.rolling_mean( df["flow_lag8"], 12)
df["flow_rollingmean_lag8_16"] = pd.rolling_mean( df["flow_lag8"], 16)
df["flow_ewma_lag1_4"] = pd.ewma( df["flow_lag1"], 4)
df["flow_ewma_lag1_8"] = pd.ewma( df["flow_lag1"], 8)
df["flow_ewma_lag1_12"] = pd.ewma( df["flow_lag1"], 12)
df["flow_ewma_lag1_16"] = pd.ewma( df["flow_lag1"], 16)
df["flow_ewma_lag4_4"] = pd.ewma( df["flow_lag4"], 4)
df["flow_ewma_lag4_8"] = pd.ewma( df["flow_lag4"], 8)
df["flow_ewma_lag4_12"] = pd.ewma( df["flow_lag4"], 12)
df["flow_ewma_lag4_16"] = pd.ewma( df["flow_lag4"], 16)
df["flow_ewma_lag8_4"] = pd.ewma( df["flow_lag4"], 4)
df["flow_ewma_lag8_8"] = pd.ewma( df["flow_lag8"], 8)
df["flow_ewma_lag8_12"] = pd.ewma( df["flow_lag8"], 12)
df["flow_ewma_lag8_16"] = pd.ewma( df["flow_lag8"], 16)
#DAYs LAG VALUES
df["flow_lag1day"] = df.groupby(["Ciclo_Estacion"])["flow"].shift(94)
df["flow_lag2day"] = df.groupby(["Ciclo_Estacion"])["flow"].shift(95)
df["flow_lag3day"] = df.groupby(["Ciclo_Estacion"])["flow"].shift(96)
df["flow_lag4day"] = df.groupby(["Ciclo_Estacion"])["flow"].shift(97)
df["flow_lag5day"] = df.groupby(["Ciclo_Estacion"])["flow"].shift(98)
df["flow_lag6day"] = df.groupby(["Ciclo_Estacion"])["flow"].shift(99)
df["flow_lag7day"] = df.groupby(["Ciclo_Estacion"])["flow"].shift(100)
df["flow_lag8day"] = df.groupby(["Ciclo_Estacion"])["flow"].shift(101)
df["flow_rollingmean_lag1day_4"] = pd.rolling_mean( df["flow_lag1day"], 4)
df["flow_rollingmean_lag1day_8"] = pd.rolling_mean( df["flow_lag1day"], 8)
df["flow_rollingmean_lag1day_12"] = pd.rolling_mean( df["flow_lag1day"], 12)
df["flow_rollingmean_lag1day_16"] = pd.rolling_mean( df["flow_lag1day"], 16)
df["flow_rollingmean_lag4day_4"] = pd.rolling_mean( df["flow_lag4day"], 4)
df["flow_rollingmean_lag4day_8"] = pd.rolling_mean( df["flow_lag4day"], 8)
df["flow_rollingmean_lag4day_12"] = pd.rolling_mean( df["flow_lag4day"], 12)
df["flow_rollingmean_lag4day_16"] = pd.rolling_mean( df["flow_lag4day"], 16)
df["flow_rollingmean_lag8day_4"] = pd.rolling_mean( df["flow_lag8day"], 4)
df["flow_rollingmean_lag8day_8"] = pd.rolling_mean( df["flow_lag8day"], 8)
df["flow_rollingmean_lag8day_12"] = pd.rolling_mean( df["flow_lag8day"], 12)
df["flow_rollingmean_lag8day_16"] = pd.rolling_mean( df["flow_lag8day"], 16)
df["flow_ewma_lag1day_4"] = pd.ewma( df["flow_lag1day"], 4)
df["flow_ewma_lag1day_8"] = pd.ewma( df["flow_lag1day"], 8)
df["flow_ewma_lag1day_12"] = pd.ewma( df["flow_lag1day"], 12)
df["flow_ewma_lag1day_16"] = pd.ewma( df["flow_lag1day"], 16)
df["flow_ewma_lag4day_4"] = pd.ewma( df["flow_lag4day"], 4)
df["flow_ewma_lag4day_8"] = pd.ewma( df["flow_lag4day"], 8)
df["flow_ewma_lag4day_12"] = pd.ewma( df["flow_lag4day"], 12)
df["flow_ewma_lag4day_16"] = pd.ewma( df["flow_lag4day"], 16)
df["flow_ewma_lag8day_4"] = pd.ewma( df["flow_lag8day"], 4)
df["flow_ewma_lag8day_8"] = pd.ewma( df["flow_lag8day"], 8)
df["flow_ewma_lag8day_12"] = pd.ewma( df["flow_lag8day"], 12)
df["flow_ewma_lag8day_16"] = pd.ewma( df["flow_lag8day"], 16)
#WEEK LAG VALUES
df["month"] = df.iteration_start.apply(lambda x: x.date().month)
df["day_month"] = df.iteration_start.apply(lambda x: x.date().day)
return df
ts_log = np.log(ts) plt.plot(ts_log) moving_avg = pd.rolling_mean(ts_log, 12) plt.plot(ts_log) plt.plot(moving_avg, color='red') ts_log_moving_avg_diff = ts_log - moving_avg ts_log_moving_avg_diff.head(12) ts_log_moving_avg_diff.dropna(inplace=True) test_stationarity(ts_log_moving_avg_diff) expwighted_avg = pd.ewma(ts_log, halflife=12) plt.plot(ts_log) plt.plot(expwighted_avg, color='red') ts_log_ewma_diff = ts_log - expwighted_avg test_stationarity(ts_log_ewma_diff) ts_log_diff = ts_log - ts_log.shift() plt.plot(ts_log_diff) ts_log_diff.dropna(inplace=True) test_stationarity(ts_log_diff) from statsmodels.tsa.seasonal import seasonal_decompose decomposition = seasonal_decompose(ts_log)
data = {"Close": before_close, "High" : high, "Low" : low}
df = pandas.DataFrame(data)
# print(df)
# print(close_dataframe)
# MACD = MACD(close_dataframe, 13, 26)
n_fast = 13
n_slow = 26
EMAfast = pd.Series(pd.ewma(df['Close'], span = n_fast))
EMAslow = pd.Series(pd.ewma(df['Close'], span = n_slow))
MACD = pd.Series(EMAfast - EMAslow, name = 'MACD_' + str(n_fast) + '_' + str(n_slow))
MACDsign = pd.Series(pd.ewma(MACD, span = 9, min_periods = 8), name = 'MACDsign_' + str(n_fast) + '_' + str(n_slow))
MACDdiff = pd.Series(MACD - MACDsign, name = 'MACDdiff_' + str(n_fast) + '_' + str(n_slow))
df = df.join(MACD)
df = df.join(MACDsign)
df = df.join(MACDdiff)
print(df["MACDdiff_13_26"])
MACD_signal = df["MACDdiff_13_26"].dropna().tolist()[-1] > 0
MACD_1.append(df["MACDdiff_13_26"].dropna().tolist()[-1])
MACD_2.append( df["MACDdiff_13_26"].dropna().tolist()[-2])
print('生成日期范围:\n', pd.date_range('2012/1/4', '2012/4/6', freq='BM'))
print('生成日期范围:\n', pd.date_range('2012/3/4', '2012/4/6', freq='W-FRI'))
ts = pd.Series(np.random.randn(4),
index=pd.date_range('1/1/2000', periods=4, freq='M'))
print('时间位移:', ts.shift(2))
print('时间位移:', ts.shift(-2))
print('时间位移:', ts.shift(1, freq='3D'))
s = pd.Series(df.trade_vol.values, index=df.time)
s2 = pd.Series(df.trade_pr.values, index=df.time)
ticks = pd.DataFrame(
{
'open': df.trade_pr.values,
'high': df.s1pr.values,
'low': df.b1pr.values,
'close': df.trade_pr.values
},
index=df.time)
ms = s.resample('5min', how=sum)
print('降采样:', ms[:5])
mt = s2.resample('1min', how='ohlc', fill_method="ffill")
print('open high low close降采样:', mt[:10])
s15t = mt.resample('15s', fill_method='ffill')
print('升采样:', s15t[:10])
mean_s = pd.rolling_mean(s, 5, min_periods=1)
print('求移动均值:', mean_s[:10])
ema_s = pd.ewma(s, 60)
print('求指数移动均值:', mean_s[:10])
df = pd.read_csv('../../data/5_minutes_dump.csv') # ШОРТЫ
df.columns = ['open', 'close','low', 'high', 'volume', 'date_time', 'ex', 'typeBlockchain']
df.index = df.date_time
df = df.sort_index()
x_scaler = MinMaxScaler()
y_scaler = MinMaxScaler()
all_df = df.copy()
x = all_df[['open', 'low', 'high', 'volume']].copy()
y = all_df['close'].copy()
x = pd.ewma(x,2)
y = pd.ewma(y,2)
x[['open', 'low', 'high', 'volume']] = x_scaler.fit_transform(x)
y = y_scaler.fit_transform(y.values.reshape(-1, 1))
shape = x.shape[1]
#X_train, y_train = load.load_data(x, WINDOW, TrainTest = False) # не удалять, чтобы переключить на сбор данных только трейна
X_train, y_train, X_test, y_test = load.load_data(x, WINDOW, train_size= 0.96, TrainTest = True)
model = build_model(input_shape=(WINDOW, shape))
print('START FIT MODEL...')
def ewma(series, span):
ema = pd.ewma(series, span=span)
return ema
yo = y_obs[0:i]
mu, var = gu.gp_solve(xo, yo, x_pred, kernel_used, sig_noise, **params)
gu.gp_plot(xo, yo, x_pred, mu, var)
plt.plot(x, fx, 'orange', alpha=0.5)
plt.pause(0.1)
# ----------------------
# Real Exmaple
#tmp = pd.read_csv('shop.csv')
tmp = pd.read_csv('auto.csv')
tmp = tmp.sort_index(ascending=False)
dt_raw = pd.Series(tmp['Close'].values, index=pd.to_datetime(tmp['Date']))
dt_smth = pd.ewma(dt_raw, halflife=15)
dt = dt_smth.resample('W-MON', how='last')
#dt = dt_smth.resample('M', how='last')
plt.figure()
plt.plot(dt, color='orange')
plt.grid(alpha=0.2)
t = np.arange(1, len(dt) + 1)
no_sample = 2
sig_noise = np.sqrt(0.01)
# periodicic
kernel_used = kf.kfunc_per
params = {'l': 5, 'p': 12 * 4}
t0 = t[0:no_sample]
y0 = dt[0:no_sample]
df = df.reset_index()
df['30 mavg'] = pd.rolling_mean(df['Close'], 30)
df['26 ema'] = pd.ewma(df['Close'], span=26)
df['12 ema'] = pd.ewma(df['Close'], span=12)
df['MACD'] = (df['12 ema'] - df['26 ema'])
df['Signal'] = pd.ewma(df['MACD'], span=9)
df['Crossover'] = df['MACD'] - df['Signal']
return stock, df['Date'][-1:].to_string(),df['Crossover'][-1:].mean()
directory = 'C:/Users/Richard Hardis/.spyder-py3/'
df = pd.read_csv(directory+'SPY.csv')
# MACD 3,6,2
df['stock_df_3_ema'] = pd.ewma(df['Close'], span=3)
df['stock_df_6_ema'] = pd.ewma(df['Close'], span=6)
df['stock_df_macd_3_6'] = df['stock_df_3_ema'] - df['stock_df_6_ema']
df['stock_df_signal_3_6'] = pd.ewma(df['stock_df_macd_3_6'], span=2)
df['stock_df_crossover_3_6'] = df['stock_df_macd_3_6'] - df['stock_df_signal_3_6'] # means, if this is > 0, or stock_df['Crossover'] = stock_df['MACD'] - stock_df['Signal'] > 0, there is a buy signal
# means, if this is < 0, or stock_df['Crossover'] = stock_df['MACD'] - stock_df['Signal'] < 0, there is a sell signal
# MACD 5,15,3
df['stock_df_5_ema'] = pd.ewma(df['Close'], span=5)
df['stock_df_15_ema'] = pd.ewma(df['Close'], span=15)
df['stock_df_macd_5_15'] = df['stock_df_5_ema'] - df['stock_df_15_ema']
df['stock_df_signal_5_15'] = pd.ewma(df['stock_df_macd_5_15'], span=3)
df['stock_df_crossover_5_15'] = df['stock_df_macd_5_15'] - df['stock_df_signal_5_15'] # means, if this is > 0, or stock_df['Crossover'] = stock_df['MACD'] - stock_df['Signal'] > 0, there is a buy signal
# means, if this is < 0, or stock_df['Crossover'] = stock_df['MACD'] - stock_df['Signal'] < 0, there is a sell signal
def __init__(self,
symbol='SPY',
sd=dt.datetime(2008, 1, 1),
ed=dt.datetime(2009, 1, 1),
sv=10000,
verbose=False):
self.symbol = symbol
self.sv = sv
self.verbose = verbose
self.cash = sv
self.shares = 0
self.position = 0
# Read data
dates = pd.date_range(sd - dt.timedelta(100), ed)
df = get_data([symbol], dates)[symbol]
# Normalize close to starting date
norm_val = get_data([symbol],
pd.date_range(sd, sd + dt.timedelta(10)))[symbol]
normed = df / norm_val.ix[0]
# Determine features
# Daily Returns
rets = pd.DataFrame(df)
daily_ret = rets[symbol].copy()
daily_ret[1:] = (rets[symbol].ix[1:] / rets[symbol].ix[:-1].values) - 1
daily_ret.ix[0] = 0
# Relative Strnength Index
up, down = daily_ret.copy(), daily_ret.copy()
up[up < 0] = 0
down[down > 0] = 0
roll_up = up.rolling(14).mean()
roll_down = down.rolling(14).mean().abs()
rsi = 100.0 - (100.0 / (1.0 + roll_up / roll_down))
# Simple moving average
sma5 = normed.rolling(5).mean()
sma10 = normed.rolling(10).mean()
sma15 = normed.rolling(15).mean()
sma20 = normed.rolling(20).mean()
sma25 = normed.rolling(25).mean()
sma30 = normed.rolling(30).mean()
sma40 = normed.rolling(40).mean()
# Volatility
vol5 = normed.rolling(5).std()
vol10 = normed.rolling(10).std()
vol20 = normed.rolling(20).std()
vol30 = normed.rolling(30).std()
# Bollinger bands
sma_bb = normed.rolling(5).mean()
sma_bb_std = normed.rolling(5).std()
bb = (normed -
(sma_bb - 2 * sma_bb_std)) / ((sma_bb + 2 * sma_bb_std) -
(sma_bb - 2 * sma_bb_std))
# Moving average convergence/divergence
ema12 = pd.ewma(np.asarray(normed), span=12)
ema26 = pd.ewma(np.asarray(normed), span=26)
macd = ema12 - ema26
# Momentum
momentum2 = normed / normed.shift(2) - 1
momentum5 = normed / normed.shift(5) - 1
momentum10 = normed / normed.shift(10) - 1
# Combine into new dataframe
df = pd.DataFrame(df).assign(sma15=normed / sma15 - 1).assign(
sma5=normed / sma5 - 1).assign(bb=bb).assign(rsi=rsi).assign(
momentum2=momentum2).assign(normed=normed).assign(
macd=macd).assign(vol10=vol10).assign(vol20=vol20).assign(
vol30=vol30).assign(vol10=vol10).assign(
sma10=normed / sma10 -
1).assign(sma20=normed / sma20 - 1).assign(
sma25=normed / sma25 -
1).assign(sma30=normed / sma30 - 1).assign(
sma40=normed / sma40 -
1).assign(vol5=vol5).assign(
momentum5=momentum5).assign(
momentum10=momentum10)[sd:]
daily_ret.ix[0] = 0
df = df.assign(dr=daily_ret)
# Determine optimal features for states
corr_df = df.corr().abs()
# Plot results
if verbose:
fig, ax = plt.subplots(figsize=(len(corr.columns),
len(corr.columns)))
ax.matshow(corr)
cmap = cm.get_cmap('coolwarm', 30)
cax = ax.imshow(df.corr(), interpolation="nearest", cmap=cmap)
plt.xticks(range(len(corr.columns)), corr.columns)
plt.yticks(range(len(corr.columns)), corr.columns)
fig.colorbar(cax, ticks=[0, .25, .5, .75, 1])
plt.show()
corr = corr_df['dr'][:]
icorr = np.asarray(corr)
scorr = icorr.argsort(
)[-4:][::-1] # select top 3 features and daily returns
scorr = scorr[1:] # remove daily returns from possible features
optimal_ftrs = []
for i in scorr:
optimal_ftrs.append(corr_df.columns[i])
self.optimal_ftrs = optimal_ftrs
self.df = df
self.market = df.iterrows()
self.current = self.market.next()
self.action = self.Action()
def calcAllInstructor(stockCode):
#MACD params is 5,10,7
pa = os.getcwd()
if os.path.exists(pa + '/' + stockCode + '_week.csv') == False:
print stockCode, " week data not exists"
return
stock_data = pandas.read_csv(stockCode + '_week.csv')
if stock_data['close'].count() == 0:
print stockCode, " data empty, maybe exist stock market"
return
stock_data['amplitude'] = numpy.round(stock_data['close'].diff(), 3)
stock_data['percent'] = numpy.round(stock_data['close'].pct_change(), 4)
stock_data['MA3'] = numpy.round(
pandas.rolling_mean(stock_data['close'], 3), 3)
stock_data['EMA5'] = numpy.round(pandas.ewma(stock_data['close'], span=5),
3)
stock_data['EMA10'] = numpy.round(
pandas.ewma(stock_data['close'], span=10), 3)
stock_data['EMA20'] = numpy.round(
pandas.ewma(stock_data['close'], span=20), 3)
stock_data['EMA22'] = numpy.round(
pandas.ewma(stock_data['close'], span=22), 3)
stock_data['DIFF'] = map(lambda x, y: x - y, stock_data['EMA10'],
stock_data['EMA22'])
stock_data['DEA'] = numpy.round(pandas.rolling_mean(stock_data['DIFF'], 7),
3)
stock_data['abspercent'] = map(lambda x: x * 100, stock_data['percent'])
stock_data['EMA15PER'] = numpy.round(
pandas.ewma(stock_data['abspercent'], span=15), 3)
#now KDJ params is 9,3,3
stock_data['low9'] = pandas.rolling_min(stock_data['close'], 9)
stock_data['high9'] = pandas.rolling_max(stock_data['close'], 9)
stock_data['quick_k'] = map(
lambda x, y, z: numpy.round((x - y) / (z - y) * 100, 2)
if z > y else 0, stock_data['close'], stock_data['low9'],
stock_data['high9'])
stock_data['quick_d'] = numpy.round(
pandas.rolling_mean(stock_data['quick_k'], 3), 2)
stock_data['slow_d'] = numpy.round(
pandas.rolling_mean(stock_data['quick_d'], 3), 2)
#calc low open diff
stock_data['maxclosediff'] = map(lambda x: x
if x > 0 else 0, stock_data['amplitude'])
stock_data['absclosediff'] = map(lambda x: x
if x > 0 else -x, stock_data['amplitude'])
stock_data['fmrsi'] = numpy.round(
pandas.ewma(stock_data['maxclosediff'], span=5), 3)
stock_data['fzrsi'] = numpy.round(
pandas.ewma(stock_data['absclosediff'], span=5), 3)
stock_data['rsi5'] = numpy.round(
map(lambda x, y: x / y * 100, stock_data['fmrsi'],
stock_data['fzrsi']), 2)
stock_data['bollstd'] = numpy.round(
pandas.rolling_std(stock_data['close'], 20), 3)
stock_data['jd'] = map(lambda x, y: x * y, stock_data['amplitude'],
stock_data['volume'])
stock_data['ejd'] = pandas.ewma(stock_data['jd'], span=3)
stock_data.to_csv(stockCode + '_weekewma.csv')
def exponential_moving_average(df, windowsize=60):
# df['ewma'] = pd.ewma(df["avg"], span=60, freq="D")
pd.ewma(df["Adj Close"], span=windowsize, freq="D")
# to remove the presence of noise from the TS we can use some smoothing techniques moving_avg = pd.rolling_mean(ts_log,8) plt.plot(ts_log) plt.plot(moving_avg, color='red') ts_log_moving_avg_diff = ts_log - moving_avg ts_log_moving_avg_diff.head(8) ts_log_moving_avg_diff.dropna(inplace=True) test_stationarity(ts_log_moving_avg_diff) # exponentially moving average expwighted_avg = pd.ewma(ts_log, halflife=8) # halflife is used to define amount of exponential decay plt.plot(ts_log) plt.plot(expwighted_avg, color='red') ts_log_ewma_diff = ts_log - expwighted_avg test_stationarity(ts_log_ewma_diff) ## here t-stat val is lesser than 1% critical value hence this gives a better result ## ELIMINATING TREND and SEASONALITY #differencing ts_log_diff = ts_log - ts_log.shift() # first order differencing plt.plot(ts_log_diff)
for key, value in dftest[4].items():
dfoutput['Critical Value (%s)' % key] = value
print dfoutput
test_stationarity(ts)
plt.show()
plt.plot(np.log(ts))
moving_avg = pd.rolling_mean(np.log(ts), 12)
plt.plot(moving_avg)
ts_log_moving_avg_diff = np.log(ts) - moving_avg
plt.show()
plt.plot(ts_log_moving_avg_diff)
ts_log_moving_avg_diff.dropna(inplace=True)
test_stationarity(ts_log_moving_avg_diff)
expwmavg = pd.ewma(np.log(ts), halflife=12)
ts_log_expwmavg = np.log(ts) - expwmavg
plt.show()
test_stationarity(ts_log_expwmavg)
plt.plot(ts_log_expwmavg)
plt.show()
plt.plot(np.log(ts) - np.log(ts).shift())
from statsmodels.tsa.seasonal import seasonal_decompose
decomposition = seasonal_decompose(np.log(ts))
trend = decomposition.trend
seasonal = decomposition.seasonal
resid = decomposition.resid
plt.show()
plt.plot(trend)
def ComputeEMA(data):
#Exponential moving average
D = PD.Series(data[:, 1])
return PD.ewma(D, 0.99)
@contact: QQ:2089973054 email:[email protected] """ import pandas as pd # ========== 从原始csv文件中导入股票数据，以浦发银行sh600000为例 # 导入数据 - 注意：这里请填写数据文件在您电脑中的路径 stock_data = pd.read_csv('stock data/sh600000.csv', parse_dates=[1]) # 将数据按照交易日期从远到近排序 stock_data.sort('date', inplace=True) # ========== 计算移动平均线 # 分别计算5日、20日、60日的移动平均线 ma_list = [5, 20, 60] # 计算简单算术移动平均线MA - 注意：stock_data['close']为股票每天的收盘价 for ma in ma_list: stock_data['MA_' + str(ma)] = pd.rolling_mean(stock_data['close'], ma) # 计算指数平滑移动平均线EMA for ma in ma_list: stock_data['EMA_' + str(ma)] = pd.ewma(stock_data['close'], span=ma) # 将数据按照交易日期从近到远排序 stock_data.sort('date', ascending=False, inplace=True) # ========== 将算好的数据输出到csv文件 - 注意：这里请填写输出文件在您电脑中的路径 stock_data.to_csv('sh600000_ma_ema.csv', index=False)
except:
try:
data_index[m] = quandl.get(mkts[m], authtoken=token).Settle
except:
try:
data_index[m] = quandl.get(mkts[m], authtoken=token).Value
except:
try:
data_index[m] = quandl.get(mkts[m], authtoken=token).value
except:
try:
data_index[m] = quandl.get(mkts[m],
authtoken=token).Rate
except:
print(m)
data_pct = data_index.pct_change()
factors = pd.DataFrame()
for f in factor_map.keys():
factors[f] = data_pct[factor_map[f]].mean(axis=1)
mu = pd.ewma(factors, 260)
sd = pd.ewmstd(factors, 260)
zscores = (factors - mu) / sd
last = zscores.iloc[-2].dropna().sort_values()
last.plot(kind='barh', colormap='jet',
ylim=[-3, 3]).get_figure().savefig('zscore.png', bbox_inches='tight')
e = Email(subject='Morning Update: Factor Dashboard')
e.add_attachment('zscore.png')
e.send()
def ema(arg, n):
if n == 0:
return pd.ewma(arg, span=len(arg), min_periods=1)
else:
return pd.ewma(arg, span=n, min_periods=n)
def EWMA(data, ndays):
EMA = pd.Series(pd.ewma(data['Close'], span=ndays, min_periods=ndays - 1),
name='EWMA_' + str(ndays))
data = data.join(EMA)
return data
def do_something():
dateparse = lambda dates: pd.datetime.strptime(dates, '%Y-%m')
data = pd.read_csv('AirPassengers.csv',
parse_dates=['date'],
index_col='date',
date_parser=dateparse)
print data.head()
ts = data['Passengers']
ts_log = np.log(ts)
# 平滑处理
moving_avg = pd.Series.rolling(ts_log, 12).mean()
# plt.plot(ts_log)
# plt.plot(moving_avg, color='red')
# plt.show()
# 做差,均值和原始值的差
ts_log_moving_avg_diff = ts_log - moving_avg
adf_test(ts_log_moving_avg_diff.dropna(), 'ts_log_moving_avg_diff')
plt.close()
# 指数加权移动平均
# 前面移动平均数需要指定window, 并且对所有的数一视同仁；
# 这里采用指数加权移动平均方法，会对当前的数据加大权重，对过去的数据减小权重。
# halflife半衰期，用来定义衰减量。其他参数, 如跨度span和质心com也可以用来定义衰减。
expwighted_avg = pd.ewma(ts_log, halflife=12)
plt.plot(ts_log)
plt.plot(expwighted_avg, color='red')
plt.savefig('pic/expwighted_avg.png')
ts_log_ewma_diff = ts_log - expwighted_avg
adf_test(ts_log_ewma_diff, 'ts_log_ewma_diff')
# Test Statistic - 3.601262
# p-value 0.005737
# #Lags Used 13.000000
# Number of Observations Used 130.000000
# Critical Value (5%) -2.884042
# Critical Value (1%) -3.481682
# Critical Value (10%) -2.578770
# 可以发现，经过指数移动平均后，再做差的结果，已经能够通过1%显著性水平检验了。
# 步长为1的一阶差分
plt.close()
ts_log_diff = ts_log - ts_log.shift(periods=1)
ts_log_diff.dropna(inplace=True)
# adf_test(ts_log_diff, 'ts_log_diff')
# 只通过了10%的显著性检验
# 我们继续进行二阶差分
# 一阶差分：Y(k)=X(k+1)-X(k)
# 二阶差分：Y(k)的一阶差分Z(k)=Y(k+1)-Y(k)=X(k+2)-2*X(k+1)+X(k)为此函数的二阶差分
ts_log_diff = ts_log - ts_log.shift(periods=1)
ts_log_diff2 = ts_log_diff - ts_log_diff.shift(periods=1)
plt.plot(ts_log_diff2)
ts_log_diff2.dropna(inplace=True)
adf_test(ts_log_diff2, 'ts_log_diff2')
# 可以看到，二阶差分，p值非常小，小于1%，检验统计量也明显小于%1的临界值。因此认定为很平稳
decompostion = seasonal_decompose(ts_log)
trend = decompostion.trend
seasonal = decompostion.seasonal
residual = decompostion.resid
plt.subplot(411)
plt.plot(ts_log, label='Original')
plt.legend(loc='best')
plt.subplot(412)
plt.plot(trend, label='Trend')
plt.legend(loc='best')
plt.subplot(413)
plt.plot(seasonal, label='Seasonality')
plt.legend(loc='best')
plt.subplot(414)
plt.plot(residual, label='Residuals')
plt.legend(loc='best')
plt.tight_layout()
plt.savefig('pic/decompostion.png')
# 对残差进行ADF检验
# 可以发现序列非常平稳
plt.close()
ts_log_decompose = residual
ts_log_decompose.dropna(inplace=True)
adf_test(ts_log_decompose, 'ts_log_decompose')
# 对残差进行ADF检验，可以发现序列非常平稳。
# 前面我们对数据进行ADF检验，判断序列是否平稳，这里我们使用自相关图和偏自相关图对数据平稳性再次进行验证，一阶差分如下图：
acf_pacf_plot(ts_log_diff) # 调用一阶差分
"""
base_capital = DEFAULT_CAPITAL
daily_risk_capital = DEFAULT_CAPITAL * DEFAULT_ANN_RISK_TARGET / ROOT_BDAYS_INYEAR
ts_capital = pd.Series(np.ones(len(price)) * DEFAULT_CAPITAL,
index=price.index)
ann_risk = ts_capital * DEFAULT_ANN_RISK_TARGET
daily_returns_volatility = robust_vol_calc(price.diff())
multiplier = daily_risk_capital * 1.0 * 1.0 / 10.0
numerator = forecast * multiplier
positions = numerator.ffill() / daily_returns_volatility.ffill()
cum_trades = positions.shift(1).ffill()
price_returns = price.diff()
instr_ccy_returns = cum_trades.shift(1) * price_returns
instr_ccy_returns = instr_ccy_returns.cumsum().ffill().reindex(
price.index).diff()
mean_return = instr_ccy_returns.mean() * BUSINESS_DAYS_IN_YEAR
vol = instr_ccy_returns.std() * ROOT_BDAYS_INYEAR
return mean_return / vol
if __name__ == "__main__":
f = '../sysdata/legacycsv/EDOLLAR_price.csv'
df = pd.read_csv(f, index_col=0, parse_dates=True)
fast_ewma = pd.ewma(df.PRICE, span=32)
slow_ewma = pd.ewma(df.PRICE, span=128)
raw_ewmac = fast_ewma - slow_ewma
vol = robust_vol_calc(df['PRICE'].diff())
forecast = raw_ewmac / vol
print(sharpe(df.PRICE, forecast))
RS1 = roll_up1 / roll_down1 RSI1 = 100.0 - (100.0 / (1.0 + RS1)) # Calculate the SMA roll_up2 = pd.rolling_mean(up, window_length) roll_down2 = pd.rolling_mean(down.abs(), window_length) RS2 = roll_up2 / roll_down2 RSI2 = 100.0 - (100.0 / (1.0 + RS2)) #12-EMA Calculation ema12 = pd.stats.moments.ewma(close, 12) #26-EMA Calculation ema26 = pd.stats.moments.ewma(close, 26) #Signal signal = pd.ewma(close, span=9) #MACD macd = ema12 - ema26 #10-SMA Calculation sma10 = pd.rolling_mean(close, window=10) #50-SMA Calculation sma50 = pd.rolling_mean(close, window=50) #Crossover = macd - signal crossover = macd - signal #Calculate CCI tp = ((df['High'] + df['Close'] + df['Low']) / 3).dropna()
def ema(y, alpha=0.20):
'''EXPONENTIAL MOVING AVERAGE using traditional weight arg.'''
# y could be a dataframe.
s = (2 / float(alpha)) - 1
# Thus default alpha has span of 9, i.e. "9-period EWMA."
return pd.ewma(y, span=s)
def gen_from_begin_to_today_stock_macd_data(stock_base_csv_path,
ts_stock_code_file_name,
stock_everyday_csv_path,
stock_everyday_csv_name):
#加载每天的数据
every_days_stock = get_every_days_stock_data(stock_everyday_csv_path,
stock_everyday_csv_name)
#old stock code list
stock_code_list = get_all_stock_code_list(stock_base_csv_path)
# ========== 根据上一步得到的代码列表，遍历所有股票，将这些股票合并到一张表格 all_stock 中
all_stock = pd.DataFrame()
i = 0
# 遍历每个股票
for code in stock_code_list:
# 测试 5 次跳过
i += 1
if i >= 5:
#break
pass
#print code
# 从csv文件中读取该股票数据
# 注意：这里请填写数据文件在您电脑中的路径
stock_data = pd.read_csv(stock_base_csv_path + code + '.csv',
parse_dates=[2],
encoding='gbk')
# print stock_data.columns
# 选取需要的字段，去除其他不需要的字段
# 股票代码,股票名称,交易日期,新浪行业,新浪概念,新浪地域,开盘价,最高价,最低价,收盘价,后复权价,前复权价,涨跌幅,成交量,成交额,换手率,流通市值,总市值,是否涨停,是否跌停,市盈率TTM,市销率TTM,市现率TTM,市净率,MA_5,MA_10,MA_20,MA_30,MA_60,MA金叉死叉,MACD_DIF,MACD_DEA,MACD_MACD,MACD_金叉死叉,KDJ_K,KDJ_D,KDJ_J,KDJ_金叉死叉,布林线中轨,布林线上轨,布林线下轨,psy,psyma,rsi1,rsi2,rsi3
stock_data = stock_data[[
u'交易日期', u'股票代码', u'收盘价', 'MACD_DIF', 'MACD_DEA', 'MACD_MACD'
]]
stock_data.columns = [
'date', 'code', 'close', 'macd_dif', 'macd_dea', 'macd_macd'
]
# 去掉 stock 代表市场的前两个字符
trim_stock_code = stock_data['code'].map(lambda x: x[2:])
stock_data['code'] = trim_stock_code
#加入 every day 的数据
cur_code_every_data = every_days_stock[every_days_stock['code'] ==
code[2:]]
stock_data = stock_data.append(cur_code_every_data, ignore_index=True)
#排序
stock_data.sort_values(by='date', ascending=True, inplace=True)
#
ema_12 = pd.ewma(stock_data['close'], span=12)
ema_26 = pd.ewma(stock_data['close'], span=26)
stock_data['ema_12'] = ema_12
stock_data['ema_26'] = ema_26
stock_data['diff'] = ema_12 - ema_26
stock_data['dea'] = pd.ewma(stock_data['diff'], span=9)
stock_data['macd'] = 2 * (stock_data['diff'] - stock_data['dea'])
#
# 将该股票的合并到output中
all_stock = all_stock.append(stock_data.tail(1), ignore_index=True)
#新的stock
new_stock_data = get_new_stock(every_days_stock, stock_code_list)
all_new_stock = gen_new_stock_macd_data(new_stock_data)
#
# 将 新股票的合并到output中
all_stock = all_stock.append(all_new_stock, ignore_index=True)
# 去除已经退市的股票
ts_stock_code = pd.read_csv(ts_stock_code_file_name,
names=['code'],
dtype={'code': str})
all_stock = all_stock[all_stock['code'].isin(ts_stock_code['code']) ==
False]
return all_stock