python_计算股票指标

注：能使用talib包的，优先使用talib包计算；本文中指标图来自同花顺截图

使用python计算常用的股票指标，本文涉及到的指标包括：RSI、OBV、MACD、 KDJ、 SAR、 VOL、 PSY、 ARBR、 CR、 EMV、 BOLL、 TRIX、 DMA、 BIAS、 CCI、 W%R、 ROC、 DMI

导入需要的包

RSI

导入需要的包


import pandas as pd
import math
import talib

RSI

中文名：相对强弱指标

计算公式：

RSI有两种计算方法：

第一种方法：

假设A为N日内收盘价涨幅的正数之和，B为N日内收盘价涨幅的负数之和再乘以（-1），这样，A和B均为正，将A，B代入RSI计算公式，则：

RSI(N) = A ÷ （A + B） × 100

第二种方法：

RS(相对强度) = N日内收盘价涨数和之均值 ÷ N日内收盘价跌数和之均值

RSI = 100 - 100 ÷ （1+RS）


    df = pd.read_csv('E:/temp005/600660.csv',encoding='utf-8')
    # 删除停牌的数据
    df = df.loc[df['openPrice']>0].copy()
    df['openPrice'] = df['openPrice']*df['accumAdjFactor']
    df['closePrice'] = df['closePrice']*df['accumAdjFactor']
    df['highestPrice'] = df['highestPrice']*df['accumAdjFactor']
    df['lowestPrice'] = df['lowestPrice']*df['accumAdjFactor']
 
    df['rsi'] = talib.RSI(df['closePrice'], timeperiod=14)

OBV

中文名：能量潮

计算公式：

OBV = 前一天的OBV ± 当日成交量

说明：（当日收盘价高于前日收盘价，成交量定位为正值，取加号；当日收盘价低于前日收盘价，成交量定义为负值，取减号；二者相等计为0）


    df = pd.read_csv('E:/temp005/600660.csv',encoding='utf-8')
    # 删除停牌的数据
    df = df.loc[df['openPrice']>0].copy()
    df['openPrice'] = df['openPrice']*df['accumAdjFactor']
    df['closePrice'] = df['closePrice']*df['accumAdjFactor']
    df['highestPrice'] = df['highestPrice']*df['accumAdjFactor']
    df['lowestPrice'] = df['lowestPrice']*df['accumAdjFactor']
 
    df['obv'] = talib.OBV(df['closePrice'],df['turnoverVol'])

MACD

中文名：指数平滑异同平均线

计算公式：

1) 简单移动平均MA

说明：P1~PN为交易日的收盘价；N为移动平均的天数

2) 加权移动平均EMA

说明：W1~WN是权重系数，W1>W2>…>WN

3) 快速时间窗口设为12日，慢速时间窗口设为26日，DIF参数设为9日，来完整计算一遍MACD

3.1) 计算指数平滑移动平均值（EMA）

12日EMA的计算公式为：

EMA(12) = 昨日EMA(12) × 11 ÷ 13 + 今日收盘价 × 2 ÷ 13

26日EMA的计算公式为：

EMA(26) = 昨日EMA(26) × 25 ÷ 27 + 今日收盘价 × 2 ÷ 27

3.2) 计算离差值（DIF）

DIF = 今日EMA(12) – 今日EMA(26)

3.3) 计算DIF的9日DEA

根据差值计算其9日的DEA，即差值平均

今日DEA = 昨日DEA × 8 ÷ 10 + 今日DIF × 2 ÷ 10


    df = pd.read_csv('E:/temp005/600660.csv',encoding='utf-8')
    # 删除停牌的数据
    df = df.loc[df['openPrice']>0].copy()
    df['openPrice'] = df['openPrice']*df['accumAdjFactor']
    df['closePrice'] = df['closePrice']*df['accumAdjFactor']
    df['highestPrice'] = df['highestPrice']*df['accumAdjFactor']
    df['lowestPrice'] = df['lowestPrice']*df['accumAdjFactor']
 
    df['DIFF'],df['DEA'],df['MACD'] = talib.MACD(close_list,fastperiod=12,slowperiod=26,signalperiod=9)

KDJ

中文名：随机指标

计算公式：

1) 以日KDJ数值的计算为例

N日RSV = (CN – LN)÷(HN-LN) ×100

说明：CN为第N日收盘价；LN为N日内的最低价；HN为N日内的最高价，RSV值始终在1~100间波动

2) 计算K值与D值

当日K值 = 2/3 × 前一日K值 + 1/3 × 当日RSV

当日D值 = 2/3 × 前一日D值 + 1/3 × 当日K值

如果没有前一日K值与D值，则可分别用50来代替

3) 计算J值

J = 3D – 2K


    df = pd.read_csv('E:/temp005/600660.csv',encoding='utf-8')
    # 删除停牌的数据
    df = df.loc[df['openPrice']>0].copy()
    df['openPrice'] = df['openPrice']*df['accumAdjFactor']
    df['closePrice'] = df['closePrice']*df['accumAdjFactor']
    df['highestPrice'] = df['highestPrice']*df['accumAdjFactor']
    df['lowestPrice'] = df['lowestPrice']*df['accumAdjFactor']
 
    df['kdj_k'],df['kdj_d'] = talib.STOCH(df['highestPrice'],df['lowestPrice'],df['closePrice'],fastk_period=5,slowk_period=3, slowk_matype=0, slowd_period=3, slowd_matype=0)
    df['kdj_j'] = 3*df['kdj_k'] - 2*df['kdj_d']

SAR

中文名：抛物转向

计算过程：

1）先选定一段时间判断为上涨或下跌

2）如果是看涨，则第一天的SAR值必须是近期内的最低价；如果是看跌，则第一天的SAR值必须是近期的最高价。

3）第二天的SAR值，则为第一天的最高价（看涨时）或是最低价（看跌时）与第一天的SAR值的差距乘上加速因子，再加上第一天的SAR值就可以求得。

4）每日的SAR值都可用上述方法类推，公式归纳如下：

SAR(N) = SAR(N-1) + AF × [(EP(N-1) – SAR(N-1))]

SAR(N) = 第N日的SAR值

SAR(N-1) = 第(N-1)日的SAR值

说明：AF表示加速因子；EP表示极点价，如果是看涨一段期间，则EP为这段时间的最高价，如果是看跌一段期间，则EP为这段时间的最低价；EP(N-1)等于第(N-1)日的极点价

5）加速因子第一次取0.02，假若第一天的最高价比前一天的最高价还高，则加速因子增加0.02，如无新高则加速因子沿用前一天的数值，但加速因子最高不能超过0.2。反之，下跌也类推

6）如果是看涨期间，计算出某日的SAR值比当日或前一日的最低价高，则应以当日或前一日的最低价为某日之SAR值；如果是看跌期间，计算某日的SAR值比当日或前一日的最高价低，则应以当日或前一日的最高价为某日的SAR值。

7）SAR指标基准周期的参数为2，如2日、2周、2月等，其计算周期的参数变动范围为2~8

8）SAR指标在股价分析系统的主图上显示为“O”形点状图。


    df = pd.read_csv('E:/temp005/600660.csv',encoding='utf-8')
    # 删除停牌的数据
    df = df.loc[df['openPrice']>0].copy()
    df['openPrice'] = df['openPrice']*df['accumAdjFactor']
    df['closePrice'] = df['closePrice']*df['accumAdjFactor']
    df['highestPrice'] = df['highestPrice']*df['accumAdjFactor']
    df['lowestPrice'] = df['lowestPrice']*df['accumAdjFactor']
 
    df['sar'] = talib.SAR(df['highestPrice'],df['lowestPrice'])

VOL

中文：成交量

柱状图是成交量，两条曲线是成交量的移动平均


    df = pd.read_csv('E:/temp005/600660.csv',encoding='utf-8')
    # 删除停牌的数据
    df = df.loc[df['openPrice']>0].copy()
    df['openPrice'] = df['openPrice']*df['accumAdjFactor']
    df['closePrice'] = df['closePrice']*df['accumAdjFactor']
    df['highestPrice'] = df['highestPrice']*df['accumAdjFactor']
    df['lowestPrice'] = df['lowestPrice']*df['accumAdjFactor']
 
    df['vol5'] = talib.MA(df['turnoverVol'],timeperiod=5)
    df['vol10'] = talib.MA(df['turnoverVol'],timeperiod=10)

PSY

中文名：心理线

计算公式：

PSY(N) = A/N × 100

说明：N为天数，A为在这N天之中股价上涨的天数


    df = pd.read_csv('E:/temp005/600660.csv',encoding='utf-8')
    # 删除停牌的数据
    df = df.loc[df['openPrice']>0].copy()
    df['openPrice'] = df['openPrice']*df['accumAdjFactor']
    df['closePrice'] = df['closePrice']*df['accumAdjFactor']
    df['highestPrice'] = df['highestPrice']*df['accumAdjFactor']
    df['lowestPrice'] = df['lowestPrice']*df['accumAdjFactor']
 
    df['ext_0'] = df['closePrice']-df['closePrice'].shift(1)
    df['ext_1'] = 0
    df.loc[df['ext_0']>0,'ext_1'] = 1
    df['ext_2'] = df['ext_1'].rolling(window=12).sum()
    df['psy'] = (df['ext_2']/12.0)*100

ARBR

中文：人气和意愿指标 AR:人气指标 BR:买卖意愿指标

计算公式：

AR(N) = N日内（H-O）之和 ÷ N日内（O-L）之和 × 100

说明：H表示当天最高价；L表示当天最低价；O表示当天开盘价；N表示设定的时间参数，一般原始参数日缺省值为26日

BR(N) = N日内（H-CY）之和 ÷ N日内（CY-L）之和 × 100

说明：H表示当天最高价；L表示当天最低价；CY表示前一交易日的收盘价，N表示设定的时间参数，一般原始参数缺省值为26日


    df = pd.read_csv('E:/temp005/600660.csv',encoding='utf-8')
    # 删除停牌的数据
    df = df.loc[df['openPrice']>0].copy()
    df['openPrice'] = df['openPrice']*df['accumAdjFactor']
    df['closePrice'] = df['closePrice']*df['accumAdjFactor']
    df['highestPrice'] = df['highestPrice']*df['accumAdjFactor']
    df['lowestPrice'] = df['lowestPrice']*df['accumAdjFactor']
 
    df['h_o'] = df['highestPrice'] - df['openPrice']
    df['o_l'] = df['openPrice'] - df['lowestPrice']
    df['h_o_sum'] = df['h_o'].rolling(window=26).sum()
    df['o_l_sum'] = df['o_l'].rolling(window=26).sum()
    df['ar'] = (df['h_o_sum']/df['o_l_sum'])*100
    df['h_c'] = df['highestPrice'] - df['closePrice']
    df['c_l'] = df['closePrice'] - df['lowestPrice']
    df['h_c_sum'] = df['h_c'].rolling(window=26).sum()
    df['c_l_sum'] = df['c_l'].rolling(window=26).sum()
    df['br'] = (df['h_c_sum']/df['c_l_sum'])*100

CR

中文：带状能力线或中间意愿指标

计算过程：

1）计算中间价，取以下四种中一种，任选：

中间价 = （最高价 + 最低价）÷2

中间价 = （最高价 + 最低价 + 收盘价）÷3

中间价 = （最高价 + 最低价 + 开盘价 + 收盘价）÷4

中间价 = （2倍的开盘价 + 最高价 + 最低价）÷4

2）计算CR:

CR = N日内（当日最高价 – 上个交易日的中间价）之和 ÷ N日内（上个交易日的中间价 – 当日最低价）之和

说明：N为设定的时间周期参数，一般原始参数日设定为26日

3）计算CR值在不同时间周期内的移动平均值：这三条移动平均曲线分别为MA1 MA2 MA3,时间周期分别为5日 10日 20日


    df = pd.read_csv('E:/temp005/600660.csv',encoding='utf-8')
    # 删除停牌的数据
    df = df.loc[df['openPrice']>0].copy()
    df['openPrice'] = df['openPrice']*df['accumAdjFactor']
    df['closePrice'] = df['closePrice']*df['accumAdjFactor']
    df['highestPrice'] = df['highestPrice']*df['accumAdjFactor']
    df['lowestPrice'] = df['lowestPrice']*df['accumAdjFactor']
 
    df['m_price'] = (df['highestPrice'] + df['lowestPrice'])/2
    df['h_m'] = df['highestPrice']-df['m_price'].shift(1)
    df['m_l'] = df['m_price'].shift(1)-df['lowestPrice']
    df['h_m_sum'] = df['h_m'].rolling(window=26).sum()
    df['m_l_sum'] = df['m_l'].rolling(window=26).sum()
    df['cr'] = (df['h_m_sum']/df['m_l_sum'])*100
    df['ma1'] = talib.MA(df['cr'],timeperiod=5)
    df['ma2'] = talib.MA(df['cr'],timeperiod=10)
    df['ma3'] = talib.MA(df['cr'],timeperiod=20)

EMV

中文：简易波动指标

计算方法：

1）先计算出三个因子A B C的数值。

A = (当日最高价 + 当日最低价)÷2

B = (上个交易日最高价 + 上个交易日最低价) ÷2

C = 当日最高价 – 当日最低价

2）求出EM数值

EM = (A-B) ×C÷当日成交额

3）求出EMV数值

EMV = EM数值的N个交易日之和，N为时间周期，一般设为14日

4）求出EMV的移动平均值EMVA

EMVA = EMV的M日移动平均值，M一般设置9日


    df = pd.read_csv('E:/temp005/600660.csv',encoding='utf-8')
    # 删除停牌的数据
    df = df.loc[df['openPrice']>0].copy()
    df['openPrice'] = df['openPrice']*df['accumAdjFactor']
    df['closePrice'] = df['closePrice']*df['accumAdjFactor']
    df['highestPrice'] = df['highestPrice']*df['accumAdjFactor']
    df['lowestPrice'] = df['lowestPrice']*df['accumAdjFactor']
 
    df['a'] = (df['highestPrice']+df['lowestPrice'])/2
    df['b'] = (df['highestPrice'].shift(1)+df['lowestPrice'].shift(1))/2
    df['c'] = df['highestPrice'] - df['lowestPrice']
    df['em'] = (df['a']-df['b'])*df['c']/df['turnoverValue']
    df['emv'] = df['em'].rolling(window=14).sum()
    df['emva'] = talib.MA(df['emv'],timeperiod=9)

BOLL

中文名：布林线指标

计算公式：

中轨线 = N日的移动平均线

上轨线 = 中轨线 + 两倍的标准差

下轨线 = 中轨线 – 两倍的标准差

计算过程：

1）先计算出移动平均值MA

MA = N日内的收盘价之和÷N

2）计算出标准差MD的平方

MD的平方 = 每个交易日的（收盘价-MA）的N日累加之和的两次方 ÷ N

3）求出MD

MD = （MD的平方）的平方根

4）计算MID、UPPER、LOWER的数值

MID = (N-1)日的MA

UPPER = MID + 2×MD

LOWER = MID – 2×MD

说明：N一般原始参数日缺省值为20日


    df = pd.read_csv('E:/temp005/600660.csv',encoding='utf-8')
    # 删除停牌的数据
    df = df.loc[df['openPrice']>0].copy()
    df['openPrice'] = df['openPrice']*df['accumAdjFactor']
    df['closePrice'] = df['closePrice']*df['accumAdjFactor']
    df['highestPrice'] = df['highestPrice']*df['accumAdjFactor']
    df['lowestPrice'] = df['lowestPrice']*df['accumAdjFactor']
 
    df['upper'],df['mid'],df['lower'] = talib.BBANDS(df['closePrice'],timeperiod=20,nbdevup=2, nbdevdn=2, matype=0)

TRIX

中文名：三重指数平滑移动平均指标


    df = pd.read_csv('E:/temp005/600660.csv',encoding='utf-8')
    # 删除停牌的数据
    df = df.loc[df['openPrice']>0].copy()
    df['openPrice'] = df['openPrice']*df['accumAdjFactor']
    df['closePrice'] = df['closePrice']*df['accumAdjFactor']
    df['highestPrice'] = df['highestPrice']*df['accumAdjFactor']
    df['lowestPrice'] = df['lowestPrice']*df['accumAdjFactor']
 
    df['trix'] = talib.TRIX(df['closePrice'],timeperiod=12)
    df['trma'] = talib.MA(df['trix'],timeperiod=20)

DMA

中文名：平均线差

计算公式：

DDD(N) = N日短期平均值 – M日长期平均值

AMA(N) = DDD的N日短期平均值

计算过程：

以求10日、50日为基准周期的DMA指标为例

1）求出周期不等的两条移动平均线MA之间的差值

DDD(10) = MA10 – MA50

2）求DDD的10日移动平均数值

AMA(10) = DDD(10)÷10


    df = pd.read_csv('E:/temp005/600660.csv',encoding='utf-8')
    # 删除停牌的数据
    df = df.loc[df['openPrice']>0].copy()
    df['openPrice'] = df['openPrice']*df['accumAdjFactor']
    df['closePrice'] = df['closePrice']*df['accumAdjFactor']
    df['highestPrice'] = df['highestPrice']*df['accumAdjFactor']
    df['lowestPrice'] = df['lowestPrice']*df['accumAdjFactor']
 
    df['ma10'] = talib.MA(df['closePrice'],timeperiod=10)
    df['ma50'] = talib.MA(df['closePrice'],timeperiod=50)
    df['ddd'] = df['ma10'] - df['ma50']
    df['ama'] = talib.MA(df['ddd'],timeperiod=10)

BIAS

中文名：乖离率

计算公式：

N日BIAS = （当日收盘价 – N日移动平均价）÷N日移动平均价×100


    df = pd.read_csv('E:/temp005/600660.csv',encoding='utf-8')
    # 删除停牌的数据
    df = df.loc[df['openPrice']>0].copy()
    df['openPrice'] = df['openPrice']*df['accumAdjFactor']
    df['closePrice'] = df['closePrice']*df['accumAdjFactor']
    df['highestPrice'] = df['highestPrice']*df['accumAdjFactor']
    df['lowestPrice'] = df['lowestPrice']*df['accumAdjFactor']
 
    df['ma6'] = talib.MA(df['closePrice'],timeperiod=6)
    df['ma12'] = talib.MA(df['closePrice'],timeperiod=12)
    df['ma24'] = talib.MA(df['closePrice'],timeperiod=24)
    df['bias'] = ((df['closePrice']-df['ma6'])/df['ma6'])*100
    df['bias2'] = ((df['closePrice']-df['ma12'])/df['ma12'])*100
    df['bias3'] = ((df['closePrice']-df['ma24'])/df['ma24'])*100

CCI

中文名：顺势指标

计算过程：

CCI(N日) = （TP-MA）÷MD÷0.015

说明：TP = （最高价+最低价+收盘价）÷3；MA=最近N日收盘价的累计之和÷N；MD=最近N日（MA-收盘价）的累计之和÷N；0.015为计算系数；N为计算周期，默认为14天


    df = pd.read_csv('E:/temp005/600660.csv',encoding='utf-8')
    # 删除停牌的数据
    df = df.loc[df['openPrice']>0].copy()
    df['openPrice'] = df['openPrice']*df['accumAdjFactor']
    df['closePrice'] = df['closePrice']*df['accumAdjFactor']
    df['highestPrice'] = df['highestPrice']*df['accumAdjFactor']
    df['lowestPrice'] = df['lowestPrice']*df['accumAdjFactor']
 
    df['cci'] = talib.CCI(df['highestPrice'],df['lowestPrice'],df['closePrice'],timeperiod=14)

W%R

中文名：威廉指标

计算公式：


    df = pd.read_csv('E:/temp005/600660.csv',encoding='utf-8')
    # 删除停牌的数据
    df = df.loc[df['openPrice']>0].copy()
    df['openPrice'] = df['openPrice']*df['accumAdjFactor']
    df['closePrice'] = df['closePrice']*df['accumAdjFactor']
    df['highestPrice'] = df['highestPrice']*df['accumAdjFactor']
    df['lowestPrice'] = df['lowestPrice']*df['accumAdjFactor']
 
    n = 10
    df['h_10'] = df['highestPrice'].rolling(window=n).max()
    df['l_10'] = df['lowestPrice'].rolling(window=n).min()
    df['wr10'] = ((df['h_10']-df['closePrice'])/(df['h_10']-df['l_10']))*100
    n2 = 6
    df['h_6'] = df['highestPrice'].rolling(window=n2).max()
    df['l_6'] = df['lowestPrice'].rolling(window=n2).min()
    df['wr6'] = ((df['h_6'] - df['closePrice']) / (df['h_6'] - df['l_6'])) * 100

ROC

中文名：变动速率指标

计算过程：

1）计算出ROC数值

ROC = (当日收盘价 – N日前收盘价)÷N日前收盘价×100

说明：N一般取值为12日

2）计算ROC移动平均线（ROCMA）数值

ROCMA = ROC的M日数值之和÷M

说明：M一般取值为6日


    df = pd.read_csv('E:/temp005/600660.csv',encoding='utf-8')
    # 删除停牌的数据
    df = df.loc[df['openPrice']>0].copy()
    df['openPrice'] = df['openPrice']*df['accumAdjFactor']
    df['closePrice'] = df['closePrice']*df['accumAdjFactor']
    df['highestPrice'] = df['highestPrice']*df['accumAdjFactor']
    df['lowestPrice'] = df['lowestPrice']*df['accumAdjFactor']
 
    df['roc'] = talib.ROC(df['closePrice'],timeperiod=12)
    df['rocma'] = talib.MA(df['roc'],timeperiod=6)

DMI

中文名：趋向指标


    df = pd.read_csv('E:/temp005/600660.csv',encoding='utf-8')
    # 删除停牌的数据
    df = df.loc[df['openPrice']>0].copy()
    df['openPrice'] = df['openPrice']*df['accumAdjFactor']
    df['closePrice'] = df['closePrice']*df['accumAdjFactor']
    df['highestPrice'] = df['highestPrice']*df['accumAdjFactor']
    df['lowestPrice'] = df['lowestPrice']*df['accumAdjFactor']
 
    df['pdi'] = talib.PLUS_DI(df['highestPrice'],df['lowestPrice'],df['closePrice'],timeperiod=14)
    df['mdi'] = talib.MINUS_DI(df['highestPrice'],df['lowestPrice'],df['closePrice'],timeperiod=14)
    df['adx'] = talib.ADX(df['highestPrice'],df['lowestPrice'],df['closePrice'],timeperiod=6)
    df['adxr'] = talib.ADXR(df['highestPrice'],df['lowestPrice'],df['closePrice'],timeperiod=6)

数据

链接：https://pan.baidu.com/s/1HPkMsDDyXTEgffoAVIhbZw
提取码：h80x

相关阅读:
ABAP json解析使用引用代替预定义数据结构
 ✔ ★ 算法基础笔记（Acwing）（六）—— 贪心【java版本】
android U广播详解（二）
JavaScript遭嫌弃，“反JS”主义者兴起
 数据库应用：Windows 部署 MySQL 8.0.36
Linux C文件操作
 Nginx解决跨域问题的一些想法
 string的底层实现
 c++数据结构算法复习基础-- 3 --线性表-单向链表-笔试面试常见问题
 Linux区分文件类型，file指令，目录权限，umask掩码，共享文件，Linux中的一些有趣指令
原文地址：https://blog.csdn.net/m0_37967652/article/details/127579127