DeepSeek大模型：通达信基于主力异动关键K线因子指标量化Python实现

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt

df['MA5'] = df[


    
'C'].rolling(window=5).mean()
df['MA10'] = df['C'].rolling(window=10).mean()
df['MA20'] = df['C'].rolling(window=20).mean()
df['MA60'] = df['C'].rolling(window=60).mean()
df['EMA13'] = df['C'].ewm(span=13, adjust=False).mean()
df['EMA12'] = df['C'].ewm(span=12, adjust=False).mean()
df[


    
'EMA26'] = df['C'].ewm(span=26, adjust=False).mean()
df['DIF'] = df['EMA12'] - df['EMA26']
df['DEA'] = df['DIF'].ewm(span=9, adjust=False).mean()
df['MACD'] = (df['DIF'] - df['DEA']) * 2
df['K线实体'] = (df['C'] - df['O']).abs()
df['WINNER'] = (df['C'] > df[


    
'C'].shift(1)).astype(int)

# 计算涨停板
df['涨停板1'] = (df['C'] == df['H']) & (df['C'] / df['C'].shift(1) > 1.0975) & \
               (df['C'] / df['C'].shift(1) 1.11) & (df['FINANCE3'] == 1) & (df['C'] == df['ZTJ'])
df['涨停板2'] = (df['C'] == df[


    
'H']) & (df['C'] / df['C'].shift(1) > 1.1985) & \
               (df['C'] / df['C'].shift(1) 1.21) & (df['FINANCE3'] == 3)
df['涨停板'] = df['涨停板1'] | df['涨停板2']

# 计算双倍量
df['双倍量'] = ((df['V'] / df['MA5'].shift(1) >= 2) & (df[


    
'V'] / df['V'].shift(1) >= 1.5) | \
               (df['V'] / df['V'].shift(1) >= 2)) & (df['C'] > df['O']) & \
               (df['H'] == df['H'].rolling(window=6).max()) & (df['C'] > df['EMA13']) & \
               (df['V'] == df['V'].rolling(window=6).max())

# 计算建仓量
df['建仓量'


    
] = df['双倍量'] & (df['K线实体'] > df['K线实体'].shift(1)) & \
               (df['V'] < df['MA5']).rolling(window=2).all()

# 计算拉升量
df['拉升量'] = df['双倍量'] & ~df['建仓量'] & df['建仓量'].rolling(window=2).any()

# 计算试盘量
df['试盘量'] = df['建仓量'] | df['拉升量']

# 计算位置要求
df['位置要求'] = (df['L'


    
] / df['EMA13'] 1.05)

# 计算红线上行
df['红线上行'] = df['EMA13'] >= df['EMA13'].shift(1)

# 计算SPX0、SPLIJIN1、SPLIJIN2、SPLIJIN3
df['SPX0'] = (df['H'] / df['C'].shift(1) >= 1.03) & df['试盘量'] & (df['H'] / df['C'] > 1.005) & \
             (df['C'] / df['C'].shift(1) > 


    
1.015) & df['位置要求'] & (df['FINANCE3'] == 1) & \
             (df['XX'] >= 40)
df['SPLIJIN1'] = (df['H'] / df['C'].shift(1) >= 1.05) & df['试盘量'] & (df['H'] / df['C'] > 1.005) & \
                 (df['C'] / df['C'].shift(1) > 1.015) & df['位置要求'


    
] & \
                 ((df['FINANCE3'] == 3) | (df['FINANCE3'] == 4)) & (df['XX'] >= 40)
df['SPLIJIN2'] = df['SPX0'] | df['SPLIJIN1']
df['SPLIJIN3'] = df['SPLIJIN2'] & df['SPLIJIN2'].rolling(window=8).sum() > 0 & \
                 (df['涨停板'].rolling(window=2).sum() == 0)

# 计算试盘K线
df['试盘K线'] = df[


    
'SPLIJIN3'].rolling(window=5).sum() > 0

# 计算建仓
df['建仓'] = df['涨停板'] & df['建仓量'] & (df['涨停板'].rolling(window=3).sum() == 1)

# 计算关键K
df['关键K'] = df['试盘K线'] | df['建仓']

# 绘制关键K线
plt.figure(figsize=(14, 7))
plt.plot(df.index, df['C'], label='Close Price')
plt.scatter(df[df['关键K'


    
]].index, df.loc[df['关键K'], 'C'], color='purple', label='关键K线', zorder=5)
plt.legend()
plt.show()