def vectorized_smabreak_divergence(code='510300', tradeCycle='D'):
colNames = [
'time', 'code', 'curclose', 'sma', 'ATR', 'ATRRatio', 'ATRWZ',
'tradeImpScore', 'primImpScore'
]
filename = StrategyName + '-' + code + '-' + tradeCycle
dfSelected = pd.DataFrame([], columns=colNames)
RootDir = os.path.join(os.pardir, 'data', 'Vectorized')
filepath = os.path.join(RootDir, filename + '.xls')
code = DP.codeType(code)
startdate = '2000-01-01'
dfdata = ts.get_k_data(code, start=startdate, end='', ktype=tradeCycle)
if (len(dfdata) < 120): # 剔除交易时间少于120天的个股
print('Not Enough data for decision making.')
totaldates = dfdata['date'].values
totalcloses = dfdata['close'].values
totalhighs = dfdata['high'].values
totallows = dfdata['low'].values
macddiclist = ta.cal_macd(totaldates, totalcloses)
totalmacd, totaldif, totaldea = DP.get_macd_list(macddiclist)
i = 120
while i < len(totalcloses):
closes = totalcloses[:i]
highs = totalhighs[:i]
lows = totallows[:i]
dates = totaldates[:i]
if (not ta.smaupcrosssignal(closes, 20)):
i += 1
continue
primCycle = GetPrimaryCycle(tradeCycle)
if (stock_in_divergece(code,
primCycle,
startdate=startdate,
enddate=dates[-1])):
absdis, perdis, maprice = ta.curdistosma(closes, 20)
curATR = ta.last_atr(dates, closes, highs, lows)
ATRWZ = absdis / curATR
curClose = closes[-1]
ATRRatio = curATR / curClose
tradeImpScore = FBD.ImpluseScore(dates, closes)
curATR = round(curATR, 2)
ATRRatio = round(ATRRatio * 100, 2)
ATRWZ = round(ATRWZ, 2)
maprice = round(maprice, 2)
dataprimCycle = ts.get_k_data(code, ktype=primCycle)
primCloses = dataprimCycle['close'].values
primdates = dataprimCycle['date'].values
primImpScore = FBD.ImpluseScore(primdates, primCloses)
s = Series([
dates[-1], code, curClose, maprice, curATR, ATRRatio, ATRWZ,
tradeImpScore, primImpScore
],
index=colNames)
dfSelected = dfSelected.append(s, ignore_index=True)
else:
pass
i += 1
# dfSelected = dfSelected.sort_values(by='totalscore',ascending=False)
dfSelected = pd.DataFrame(dfSelected.values, columns=dfSelected.columns)
# dfSelected.to_csv(filepath)
DI.Write_DF_T0_Excel(filepath, dfSelected, 'selected')
return dfSelected
def batch_upcrossafter_divergence(dataset='hs300', tradeCycle='D'):
lasttradetimestr = DP.get_last_trade_time('D')
colNames = [
'code', 'name', 'period', 'curclose', 'sma', 'ATR', 'ATRRatio',
'rewardrisk', 'primImpScore', 'tradeImpScore', 'primdivscore',
'divscore', 'difdeascore', 'macdscore', 'totalscore'
]
filename = StrategyName + lasttradetimestr + '-' + dataset + '-' + tradeCycle
RootDir = os.path.join(os.pardir, 'data', lasttradetimestr)
DP.mkdir(RootDir)
filepath = os.path.join(RootDir, filename + '.xls')
codelist = DI.get_stockcode_list(dataset)
dfSelected = pd.DataFrame([], columns=colNames)
dfScreen = codelist[['code', 'name']]
totalrows = len(dfScreen)
for i in range(totalrows):
code = dfScreen.loc[i, 'code']
code = DP.codeType(code)
name = dfScreen.loc[i, 'name']
if (DI.is_tingpai(code)):
continue
dfdata = ts.get_k_data(code, ktype=tradeCycle)
if (len(dfdata) < 120): #剔除交易时间少于120天的个股
continue
dates = dfdata['date'].values
closes = dfdata['close'].values
highs = dfdata['high'].values
lows = dfdata['low'].values
if (not ta.smaupcrosssignal(closes, 20)):
continue
print(name)
macddiclist = ta.cal_macd(dates, closes)
macd, dif, dea = DP.get_macd_list(macddiclist)
if (min(macd[-3:]) > 0):
continue
primCycle = GetPrimaryCycle(tradeCycle)
dataprimCycle = ts.get_k_data(code, ktype=primCycle)
primCloses = dataprimCycle['close'].values
primdates = dataprimCycle['date'].values
primlows = dataprimCycle['low'].values
primmacd = pd.Series(ta.cal_macd(primdates, primCloses))
primmacd, primdif, primdea = DP.get_macd_list(primmacd)
divStatus = ta.macd_in_bulldivergence(dates, primmacd)
divscore = divStatus['div']
macdscore = divStatus['macd']
period = divStatus['period']
if (divscore != 0):
print(code, name, divscore, macdscore, period)
absdis, perdis, maprice = ta.curdistosma(closes, 20)
curATR = ta.last_atr(dates, closes, highs, lows)
ATRWZ = absdis / curATR
curClose = closes[-1]
ATRRatio = curATR / curClose
primImpScore = FBD.ImpluseScore(primdates, primCloses)
tradeImpScore = FBD.ImpluseScore(dates, closes)
primdivStatus = ta.macd_in_bulldivergence_signal(
primdates, primmacd)
primdivscore = primdivStatus['div']
if (primdivscore != 0):
primdivscore = 1
difdeadivscore = 0
if (ta.in_loose_bulldivergence(dif, period)):
difdeadivscore += 1
if (ta.in_loose_bulldivergence(dea, period)):
difdeadivscore += 1
totalscore = primImpScore + tradeImpScore + primdivscore + divscore + difdeadivscore + macdscore
curATR = round(curATR, 2)
ATRRatio = round(ATRRatio * 100, 2)
ATRWZ = round(ATRWZ, 2)
maprice = round(maprice, 2)
colNames = [
'code', 'name', 'period', 'curclose', 'sma', 'ATR', 'ATRRatio',
'ATRWZ', 'primImpScore', 'tradeImpScore', 'primdivscore',
'divscore', 'difdeascore', 'macdscore', 'totalscore'
]
s = Series([
code, name, period, curClose, maprice, curATR, ATRRatio, ATRWZ,
primImpScore, tradeImpScore, primdivscore, divscore,
difdeadivscore, macdscore, totalscore
],
index=colNames)
dfSelected = dfSelected.append(s, ignore_index=True)
dfSelected = dfSelected.sort_values(by='totalscore', ascending=False)
dfSelected = pd.DataFrame(dfSelected.values, columns=dfSelected.columns)
DI.Write_DF_T0_Excel(filepath, dfSelected, 'selected')