选股票法"/>
这写股票明天要涨,Python 量化分析(五) 潜力指数选股票法
本人通过过著名的金融库talib构造时间序列的数学模型,分别计算中国3518只股票的潜力指数,分析得出未来,即5月二日开盘时,极大概率上涨的有282只,大概率上涨的有577只:
在未来如下股票将上涨:
1. 002027
1. 002546
2. 600552
3. 000417
4. 002319
5. 601002
6. 000417
7. 002319
8. 601002
在未来如下股票将下跌:
1. 601185
2. 601016
3. 002692
4. 002406
# -*- coding: utf-8 -*-
"""
Created on Thu Dec 14 15:26:31 2017@author: 量化之王
"""#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Created on Wed May 2 17:28:50 2018@author: luogan
"""import pymongo
import pandasimport pandas as pd
import matplotlib.pyplot as plt
import numpy as np
import pylab as pl
import datetime
import matplotlib.pyplot as plt
from matplotlib.dates import DateFormatter, WeekdayLocator, DayLocator, MONDAY,YEARLY
from matplotlib.finance import quotes_historical_yahoo_ohlc, candlestick_ohlcfrom matplotlib.pylab import date2num
#import potentialimport talib
from dateutil.parser import parse
import tushare as tsclient1 = pymongo.MongoClient('192.168.10.182',27017)
db1 = client1.stock.potential'''
def before_month_lastday(ti):from dateutil.parser import parsetoday=parse(str(ti))#first = datetime.date(day=1, month=today.month, year=today.year)lastMonth = today - datetime.timedelta(days=0)def plus(k):if k<10:return '0'+str(k)else:return str(k)y=lastMonth.yearm=lastMonth.monthd=lastMonth.day#day=calendar.monthrange(y,m)[1]cc=str(y)+plus(m)+plus(d)#bb=parse(cc)#pacific = pytz.timezone('Asia/Shanghai')#return pacific.localize(bb) return int(cc)
'''def potential_index(tl):#df=ts.get_hist_data(name,start=bf,end=now)df=ts.get_hist_data(tl[0],start=tl[1],end=tl[2])if str(type(df))!="<class 'NoneType'>":if df.shape[0]>10:date=df.indexdate1=list(map(parse,date))df['date']=date1df=df.sort_values(by='date')print('df=',df)df.index=list(range(df.shape[0]))#df[df['volume']==0]=np.nan#print('df=',df)"""def myMACD(price, fastperiod=12, slowperiod=26, signalperiod=9):ewma12 = pd.ewma(price,span=fastperiod)ewma60 = pd.ewma(price,span=slowperiod)dif = ewma12-ewma60dea = pd.ewma(dif,span=signalperiod)bar = (dif-dea) #有些地方的bar = (dif-dea)*2,但是talib中MACD的计算是bar = (dif-dea)*1return dif,dea,bar"""#print(df['close'].values)macd, signal, hist = talib.MACD(df['close'].values, fastperiod=6, slowperiod=12, signalperiod=9)"""#mydif,mydea,mybar = myMACD(df['close'].values, fastperiod=12, slowperiod=26, signalperiod=9)fig = plt.figure(figsize=[10,5])plt.plot(df.index,macd,label='macd dif')plt.plot(df.index,signal,label='signal dea')plt.plot(df.index,hist,label='hist bar')#plt.plot(df.index,mydea,label='my dea')#plt.plot(df.index,mybar,label='my bar')plt.legend(loc='best')"""close = [float(x) for x in df['close']]def macscore( hist):span=len(macd)-1h1=hist[span]if h1>0:return 1else:return 0def RSI(df):df['RSI']=talib.RSI(np.array(close), timeperiod=12) aa=list(df['RSI'])b=aa[::-1]#print(b)if b[0]>50:return 0else:return 1def monment(df):df['MOM']=talib.MOM(np.array(close), timeperiod=5)aa=list(df['MOM'])b=aa[::-1]if b[0]>0:return 1else:return 0def polyfit(close,k,pl):#print(close)near_six=close[len(close)-pl:len(close)]xlist=list(range(pl))bbz1 = np.polyfit(xlist, near_six,k)# 生成多项式对象{bbp1 = np.poly1d(bbz1)f5=bbp1(pl-1)f6=bbp1(pl)if f6>f5:return 1else:return 0score=2*RSI(df)+2*monment(df)+3*polyfit(close,1,2)+2*polyfit(close,1,3)+1*polyfit(close,1,4)+2*polyfit(close,3,5)+2*macscore( hist)poindex=score/14vv=int(poindex*100)db1.save({'name':tl[0],'potential':vv})#return vv*1.0#mm=potential_index(code[100])'''
for name in code:mm=potential_index(name)print(name,mm)timm=datetime.datetime.now()'''
ak=ts.get_stock_basics()code=list(ak.index)def front_step_time(day):now = datetime.datetime.now()front = now - datetime.timedelta(days=day)d1 = front.strftime('%Y-%m-%d')#return int(d1)return d1now=front_step_time(0)bf=front_step_time(720)sheet=pd.DataFrame()
sheet['code']=codesheet['bf']=bf
sheet['sta']=now
#name='600354'
#b1=potential_vocanol(name,'2017-11-14','2018-02-14')
#b2=potential_vocanol(name,'2018-02-14','2018-04-13')
client1 = pymongo.MongoClient('192.168.10.182',27017)
db1 = client1.stock.potentialimport time
from multiprocessing import Pool
import numpy as npif __name__ == "__main__" :startTime = time.time()testFL =sheet.values#ll=codepool = Pool(10)#可以同时跑10个进程pool.map(potential_index,testFL)pool.close()pool.join() endTime = time.time()print ("time :", endTime - startTime)
{ "_id" : ObjectId("5ae84623a39ad471dbdd0e8a"), "name" : "000043", "potential" : NumberInt(100)
}
{ "_id" : ObjectId("5ae84623a39ad471dadd0e87"), "name" : "002689", "potential" : NumberInt(100)
}
{ "_id" : ObjectId("5ae84623a39ad471d5dd0e8d"), "name" : "601107", "potential" : NumberInt(100)
}
{ "_id" : ObjectId("5ae84624a39ad471dbdd0e8d"), "name" : "002660", "potential" : NumberInt(100)
}
{ "_id" : ObjectId("5ae84624a39ad471dcdd0e95"), "name" : "300497", "potential" : NumberInt(100)
}
{ "_id" : ObjectId("5ae84624a39ad471d7dd0e8c"), "name" : "600353", "potential" : NumberInt(100)
}
更多推荐
这写股票明天要涨,Python 量化分析(五) 潜力指数选股票法
发布评论