python macd_利用python编写macd、kdj、rsi、ma等指标.doc

利用python编写macd、kdj、rsi、ma等指标

# -*- coding: utf-8 -*-

"""

Created on Thu Dec 15 13:57:32 2016

@author: four

"""

import pandas as pd

#获取地址数据

def get_adress_data(adress=0):

data=pd.read_csv(adress,parse_dates=False,header=None,names=['dateL','openL','highL','lowL','closeL','volL'])

data['dateL']=pd.to_datetime(data.dateL)

return data

#获取macd

def get_macd_data(data,short=0,long1=0,mid=0):

if short==0:

short=12

if long1==0:

long1=26

if mid==0:

mid=9

data['sema']=pd.ewma(data['closeL'],span=short)

data['lema']=pd.ewma(data['closeL'],span=long1)

data.fillna(0,inplace=True)

data['data_dif']=data['sema']-data['lema']

data['data_dea']=pd.ewma(data['data_dif'],span=mid)

data['data_macd']=2*(data['data_dif']-data['data_dea'])

data.fillna(0,inplace=True)

return data[['data_dif','data_dea','data_macd']]

def get_kdj_data(data,N=0,M=0):

if N==0:

N=9

if M==0:

M=2

low_list = pd.rolling_min(data['lowL'], N)

low_list.fillna(value=pd.expanding_min(data['lowL']), inplace=True)

high_list = pd.rolling_max(data['highL'],N)

high_list.fillna(value=pd.expanding_max(data['highL']), inplace=True)

rsv = (data['closeL'] - low_list) / (high_list - low_list) * 100

data['KDJ_K'] = pd.ewma(rsv,com=M)

data['KDJ_D']=pd.ewma(data['KDJ_K'],com=M)

data['KDJ_J'] = 3 * data['KDJ_K'] - 2 * data['KDJ_D']

data.fillna(0,inplace=True)

return data[['KDJ_K','KDJ_D','KDJ_J']]

def get_ma_data(data,N=0):

if N==0:

N=5

data['ma']=pd.rolling_mean(data['closeL'],N)

data.fillna(0,inplace=True)

return data[['ma']]

def get_rsi_data(data,N=0):

if N==0:

N=24

data['value']=data['closeL']-data['closeL'].shift(1)

data.fillna(0,inplace=True)

data['value1']=data['value']

data['value1'][data['value1']<0]=0

data['value2']=data['value']

data['value2'][data['value2']>0]=0

data['plus']=pd.rolling_sum(data['value1'],N)

data['minus']=pd