import numpy as np import math import akshare as ak import os from datetime import datetime, timedelta, date import pandas as pd import mplfinance as mpf import sqlite3 import stock_database import mysql_database from EmailTest import send_email, parse_return_email from dataclasses import dataclass import time @dataclass class BuyState: trigger_time: float # 触发次数 buy_price: float # 买入价格 add_price: float # 加仓价格 stop_price: float # 止损价格 shares: int # 买入股数 atr: int # ATR available_cash: float # 可用资金 @dataclass class TradeLog: data: str # 时间 type: str # 操作类型 buy_price: float # 买入价格 shares: int # 买入股数 cost: float # 成本 atr: int # ATR available_cash: float # 可用资金 all_shares: float # 总股数 all_cost: float # 总成本 Net_value: float # 净值 Net_return: float # 净收益 def calc_sma_atr_pd(kdf,period): """计算TR与ATR Args: kdf (_type_): 历史数据 period (_type_): ATR周期 Returns: _type_: 返回kdf，增加TR与ATR列 """ kdf['最高'] = kdf['最高'].astype(float) kdf['最低'] = kdf['最低'].astype(float) kdf['收盘'] = kdf['收盘'].astype(float) kdf['HL'] = kdf['最高'] - kdf['最低'] kdf['HC'] = np.abs(kdf['最高'] - kdf['收盘'].shift(1)) kdf['LC'] = np.abs(kdf['最低'] - kdf['收盘'].shift(1)) kdf['TR'] = np.round(kdf[['HL','HC','LC']].max(axis=1), 3) # ranges = pd.concat([high_low, high_close, low_close], axis=1) # true_range = np.max(ranges, axis=1) kdf['ATR'] = np.round(kdf['TR'].rolling(period).mean(), 3) return kdf.drop(['HL','HC','LC'], axis = 1) class TurtleTrading(object): """对象范围较小，对某一个标的创建一个海龟，如513300， 计算ATR、唐奇安通道线 基础数据 Args: object (_type_): _description_ """ def __init__(self, TradeCode, type, riskcoe, Capital, cash) -> None: self.TradeCode = TradeCode self.type = type self.riskcoe = riskcoe self.Capital = Capital self.cash = cash self.TrigerTime = 0 self.BuyStates = list[BuyState] = [] self.tradeslog = list[TradeLog] # 交易记录 def GetRecentData(self): """获取某个标的的最近数据，从两年前到今天, 计算后的数据保存在self.CurrentData Returns: _type_: _description_ """ Today = datetime.today() # print(Today) formatted_date = Today.strftime("%Y%m%d") two_years_ago = (date.today() - timedelta(days=365*2)).strftime("%Y%m%d") # print(formatted_date) Code = f"{self.TradeCode}" CurrentData = ak.fund_etf_hist_em(symbol=Code, period="daily", start_date=two_years_ago, end_date=formatted_date, adjust="") # 将日期列转换为datetime CurrentData = pd.DataFrame(CurrentData) CurrentData['日期'] = pd.to_datetime(CurrentData['日期']) # print(type(CurrentData['日期'].iloc[0])) CurrentData.set_index('日期', inplace=True) # CurrentData.reset_index(inplace=True) # print(type(CurrentData['日期'].iloc[0])) # create table # stock_database.create_table(Code) # stock_database.insert_data(Code, CurrentData) # mysql_database.insert_db(CurrentData, Code, True, "'日期'") self.CurrentData = CurrentData # return self.CurrentData def CalATR(self, data, ATRday): """计算某个标的的ATR，从上市日到今天, 计算后的数据保存在self.CurrentData Args: ATRday: 多少日ATR SaveOrNot (_type_): 是否保存.csv数据 """ self.CurrentData = calc_sma_atr_pd(data, ATRday) self.N = self.CurrentData['ATR'] # return self.N def ReadExistData(self, data): """除了通过发请求获取数据，也可以读本地的数据库，赋值给self.CurrentData Args: data (_type_): 本地csv名称 """ self.CurrentData = pd.read_csv(data) def DrawKLine(self, days): """画出k线图看看,画出最近days天的K线图 """ # 日期部分 # dates = pd.to_datetime(self.CurrentData['日期'][-days:]) # # Klinedf['Data'] = pd.to_datetime(self.CurrentData['日期']) Klinedf = pd.DataFrame() # Klinedf.set_index = Klinedf['Data'] # 其他数据 Klinedf['Date'] = self.CurrentData['日期'][-days:] Klinedf['Open'] = self.CurrentData['开盘'][-days:].astype(float) Klinedf['High'] = self.CurrentData['最高'][-days:].astype(float) Klinedf['Low'] = self.CurrentData['最低'][-days:].astype(float) Klinedf['Close'] = self.CurrentData['收盘'][-days:].astype(float) Klinedf['Volume'] = self.CurrentData['成交量'][-days:].astype(float) Klinedf.set_index(pd.to_datetime(Klinedf['Date']), inplace=True) # 画图 mpf.plot(Klinedf, type='candle', style='yahoo', volume=False, mav=(5,), addplot=[mpf.make_addplot(self.Donchian_up['Upper'][-days:]), mpf.make_addplot(self.Donchian_down['lower'][-days:])], title=f"{self.TradeCode} K线图") def calculate_donchian_channel_up(self, n): """ 计算n日唐奇安上通道 参数: self.CurrentData (DataFrame): 包含价格数据的Pandas DataFrame，包含"High" n (int): 时间周期 返回:self.Donchian DataFrame: 唐奇安通道的DataFrame，包含"Upper" """ Donchian = pd.DataFrame() # 创建一个空的DataFrame用于存储唐奇安通道数据 # 计算最高价和最低价的N日移动平均线 name = 'Donchian_' + str(n) + '_upper' Donchian[name] = self.CurrentData['最高'].rolling(n).max() # 使用rolling函数计算n日最高价的移动最大值 # # 计算中间线 # Donchian['Middle'] = (self.Donchian['Upper'] + self.Donchian['Lower']) / 2 # 计算上通道和下通道的中间线，但此行代码被注释掉了 return Donchian # 返回包含唐奇安上通道的DataFrame def calculate_donchian_channel_down(self, n): """ 计算n日唐奇安上通道 参数: self.CurrentData (DataFrame): 包含价格数据的Pandas DataFrame，包含"High" n (int): 时间周期 返回:self.Donchian DataFrame: 唐奇安通道的DataFrame，包含"Upper" """ Donchian = pd.DataFrame() # 计算最高价和最低价的N日移动平均线 name = 'Donchian_' + str(n) + '_lower' Donchian[name] = self.CurrentData['最低'].rolling(n).min() # # 计算中间线 # Donchian['Middle'] = (self.Donchian['Upper'] + self.Donchian['Lower']) / 2 return Donchian def calc_atr_donchian_short(self): """计算ATR、短期唐奇安通道 """ # 计算ATR self.CalATR(self.CurrentData, 20) # 计算唐奇安通道 self.Donchian_20_ups = self.calculate_donchian_channel_up(20) self.Donchian_50_ups = self.calculate_donchian_channel_up(50) self.Donchian_downs = self.calculate_donchian_channel_down(10) # 画图 # self.DrawKLine(days) # 把self.N, self.Donchian_up, self.Donchian_down, 添加到self.CurrentData后面，保存到mysql数据库 self.CurrentData = pd.concat([self.CurrentData, self.Donchian_20_ups, self.Donchian_50_ups, self.Donchian_downs], axis=1) def get_ready(self, days): """创建一个turtle对象，获取数据，计算ATR，计算唐奇安通道 Args: days (_type_): _description_ n (_type_): _description_ """ # if 不存在database if not mysql_database.check_db_table(f"{self.TradeCode}"): self.GetRecentData() self.calc_atr_donchian_short() Code = f"{self.TradeCode}" mysql_database.insert_db(self.CurrentData, Code, True, "日期") else: # 检查数据库最后一条的时间距离今天是否两天以上 current_date = date.today() threshold_date = current_date - timedelta(days=2) last_update = mysql_database.check_db_table_last_date(f"{self.TradeCode}") if last_update < threshold_date: # 如果不存在，则从akshare获取数据并保存到mysql数据库 mysql_database.delete_table(f"{self.TradeCode}") self.GetRecentData() self.calc_atr_donchian_short() Code = f"{self.TradeCode}" mysql_database.insert_db(self.CurrentData, Code, True, "日期") else: # 如果存在，则从mysql数据库中读取数据 self.CurrentData = mysql_database.fetch_all_data(f"{self.TradeCode}") def CalPositionSize(self): """根据风险系数、ATR，计算仓位大小, 存于self.IntPositionSize """ PositionSize = self.riskcoe * self.Capital /(self.N) # 默认用股票形式了 100 self.IntPositionSize = int(PositionSize // 100) * 100 def system1EnterNormal(self, PriceNow, TempDonchian20Upper, BreakOutLog): # 没有持仓且价格向上突破---此时包含两种情形：1 对某标的首次使用系统，2 已发生过突破，此时上次突破天然是失败的 if self.TrigerTime == 0 and PriceNow > TempDonchian20Upper: # 买入 return True elif PriceNow > TempDonchian20Upper:#todo !=0不会满足条件 先跳过 self.system1BreakoutValid(PriceNow) if BreakOutLog[-1][5] == 'Lose': # TT!= 0且突破且上一次突破unseccessful return True else: return False else: return False def system1EnterSafe(self, PriceNow, TempDonchian55Upper): if PriceNow > TempDonchian55Upper[-1]: # 保底的55日突破 return True else: return False def system1BreakoutValid(self, priceNow): """判断前一次突破是否成功，是log[-1][5]写入“win”，否则写入“Lose” """ if priceNow < self.BreakOutLog[-1][3]: self.BreakOutLog[-1][5] = 'Lose' else: self.BreakOutLog[-1][5] = 'None' # 一天结束，计算ATR，计算唐奇安通道，追加到已有的mysql数据库中 def system_1_Out(self, PriceNow, TempDonchian10Lower): # 退出:低于20日最低价（多头方向）,空头以突破20日最高价为止损价格--有持仓且价格向下突破 if self.TrigerTime != 0 and PriceNow < TempDonchian10Lower: # 退出 return True else: return False def add(self, PriceNow): """加仓 """ if self.TrigerTime < 4 and PriceNow > self.BuyStates[self.TrigerTime - 1][2]:#todo BuyStates是空的 # 买入 return True else: return False def system_1_stop(self, PriceNow): """止损判断：如果当前价格<上一次买入后的止损价格则止损 """ if PriceNow < self.BuyStates[self.TrigerTime - 1][3]: # 买入 return True else: return False def day_end(self): pass class TurtleTrading_OnTime(object): ''' 实时监测主程序，可以处理多个turtle 1、获取实时大盘数据 2、根据turtles的代码，比较是否触发条件 3、实时监测主流程 ''' def __init__(self, turtle: TurtleTrading, user_email): self.turtle = turtle self.user_email = user_email def get_stocks_data(self): """获取实时股票、基金数据，不保存 """ stock_data = ak.stock_zh_a_spot_em() stock_data = stock_data.dropna(subset=['最新价']) # # print(stock_zh_a_spot_df) # # stock_zh_a_spot_df第一列加上时间，精确到分钟 # stock_zh_a_spot_df['时间'] = datetime.now().strftime("%Y-%m-%d %H:%M:%S") # mysql_database.insert_db(stock_zh_a_spot_df, "stock_price", True, "代码") # etf_data = ak.fund_etf_spot_em() etf_data = ak.fund_etf_spot_ths() etf_data = etf_data.dropna(subset=['当前-单位净值']) # etf_data['时间'] = datetime.now().strftime("%Y-%m-%d %H:%M:%S") # mysql_database.insert_db(etf_data, "etf_price", True, "代码") return stock_data, etf_data def Buy_stock(self, price_now): # 发送邮件 代码self.turtle.TradeCode, 建议买入价格price_now，买入份额self.turtle.IntPositionSize if self.turtle.TrigerTime == 0: # 第一次买入 subject = "买入" body = f"{self.turtle.TradeCode}，价格{price_now}，份额{self.turtle.IntPositionSize} \n " body += "回复：实际买入价格-买入份额-手续费" send_email(subject, body, self.user_email) send_email_time = datetime.now().strftime("%Y-%m-%d %H:%M:%S") #每隔1分钟检测回信，解析邮件。 parsed_email_flag = False while not parsed_email_flag: time.sleep(60) # 每次尝试前等待 60 秒 parse_states, buy_price, buy_share, fee = parse_return_email( self.user_email, send_email_time ) if parse_states: parsed_email_flag = True break # 成功买入 self.turtle.TrigerTime += 1 # 记录self.turtle.BuyStates add_price = buy_price + 1/2 * self.turtle.N stop_price = buy_price - 2 * self.turtle.N cost = buy_price * buy_share - fee available_cash = self.turtle.Capital - cost buy_this_time = BuyState(self.turtle.TrigerTime, buy_price, add_price, stop_price, buy_share, self.turtle.N, available_cash) self.turtle.BuyStates.append(buy_this_time) today = datetime.now().strftime("%Y-%m-%d") log_this_time = TradeLog(today, "买入", buy_price, buy_share, cost, self.turtle.N, available_cash, all_shares=buy_share, all_cost=cost, Net_value=buy_price * buy_share, Net_return=0) self.turtle.tradeslog.append(log_this_time) else: # 加仓 subject = "加仓" body = f"{self.turtle.TradeCode}，价格{price_now}，份额{self.turtle.IntPositionSize} \n " body += "回复：实际买入价格-买入份额-手续费" send_email(subject, body, self.user_email) send_email_time = datetime.now().strftime("%Y-%m-%d %H:%M:%S") #每隔1分钟检测回信，解析邮件。 parsed_email_flag = False while not parsed_email_flag: time.sleep(60) # 每次尝试前等待 60 秒 parse_states, buy_price, buy_share, fee = parse_return_email( self.user_email, send_email_time ) if parse_states: parsed_email_flag = True break # 成功买入 self.turtle.TrigerTime += 1 # 记录self.turtle.BuyStates add_price = buy_price + 1/2 * self.turtle.N stop_price = buy_price - 2 * self.turtle.N cost = buy_price * buy_share - fee available_cash = self.turtle.BuyStates[-1].available_cash - cost all_shares = buy_share + self.turtle.BuyStates[-1].all_shares all_cost = cost + self.turtle.BuyStates[-1].all_cost net_value = buy_price * all_shares net_return = net_value - all_cost buy_this_time = BuyState(self.turtle.TrigerTime, buy_price, add_price, stop_price, buy_share, self.turtle.N, available_cash) self.turtle.BuyStates.append(buy_this_time) today = datetime.now().strftime("%Y-%m-%d") log_this_time = TradeLog(today, "加仓", buy_price, buy_share, cost, self.turtle.N, available_cash, all_shares, all_cost, net_value, net_return) self.turtle.tradeslog.append(log_this_time) pass def stop_sale_stock(self, price_now): """止损卖出 Args: price_now (_type_): 现价 """ # 发送邮件 代码self.turtle.TradeCode, 建议卖出价格price_now，卖出份额self.turtle.IntPositionSize subject = "止损卖出" body = f"{self.turtle.TradeCode}，价格{price_now}，份额{self.turtle.IntPositionSize} \n " body += "回复：实际卖出价格-卖出份额-手续费" send_email(subject, body, self.user_email) send_email_time = datetime.now().strftime("%Y-%m-%d %H:%M:%S") # 每隔1分钟检测回信，解析邮件。 parsed_email_flag = False while not parsed_email_flag: time.sleep(60) # 每次尝试前等待 60 秒 parse_states, sale_price, sale_share, fee = parse_return_email( self.user_email, send_email_time ) if parse_states: parsed_email_flag = True break # 成功卖出 self.turtle.TrigerTime = 0 # 记录self.turtle.BuyStates available_cash = self.turtle.BuyStates[-1].available_cash + sale_price * sale_share - fee self.turtle.BuyStates = [] sale_this_time = TradeLog(datetime.now().strftime("%Y-%m-%d"), "止损", sale_price, sale_share, sale_price * sale_share - fee, self.turtle.N, available_cash, all_shares=0, all_cost=0, Net_value=sale_price * sale_share, Net_return=abs(self.turtle.Capital - available_cash)) self.turtle.tradeslog.append(sale_this_time) def out_sale_stock(self, price_now): """止盈卖出 Args: price_now (_type_): 现价 """ # 发送邮件 代码self.turtle.TradeCode, 建议卖出价格price_now，卖出份额self.turtle.IntPositionSize subject = "止盈卖出" body = f"{self.turtle.TradeCode}，价格{price_now}，份额{self.turtle.IntPositionSize} \n " body += "回复：实际卖出价格-卖出份额-手续费" send_email(subject, body, self.user_email) send_email_time = datetime.now().strftime("%Y-%m-%d %H:%M:%S") # 每隔1分钟检测回信，解析邮件。 parsed_email_flag = False while not parsed_email_flag: time.sleep(60) # 每次尝试前等待 60 秒 parse_states, sale_price, sale_share, fee = parse_return_email( self.user_email, send_email_time ) if parse_states: parsed_email_flag = True break # 成功卖出 self.turtle.TrigerTime = 0 # 记录self.turtle.BuyStates available_cash = self.turtle.BuyStates[-1].available_cash + sale_price * sale_share - fee self.turtle.BuyStates = [] sale_this_time = TradeLog(datetime.now().strftime("%Y-%m-%d"), "止盈", sale_price, sale_share, sale_price * sale_share - fee, self.turtle.N, available_cash, all_shares=0, all_cost=0, Net_value=sale_price * sale_share, Net_return=abs(self.turtle.Capital - available_cash)) self.turtle.tradeslog.append(sale_this_time) def run_short_trading_loop(self): while True: # 获取当前时间 now = datetime.now().time() # 判断当前时间是否在交易时段内（9:30-11:30 或 13:00-15:00） is_trading_time = ( (now.hour == 9 and now.minute >= 30) or (now.hour == 10 and 0 <= now.minute <= 59) or (now.hour == 11 and now.minute <= 30) or (now.hour == 13 and 0 <= now.minute <= 59) or (now.hour == 14 and 0 <= now.minute <= 59) or (now.hour == 15 and now.minute <= 0) ) if not is_trading_time: # 非交易时间，等待 1 分钟后继续循环 time.sleep(60) continue # 获取股票和ETF数据 stock_data, etf_data = self.get_stocks_data() # 根据类型获取当前价格 if self.turtle.Type == "stock": self.turtle.PriceNow = stock_data[ stock_data['代码'] == self.turtle.TradeCode ]['最新价'].iloc[-1] elif self.turtle.Type == "etf": self.turtle.PriceNow = etf_data[ etf_data['代码'] == self.turtle.TradeCode ]['当前-单位净值'].iloc[-1] # 判断当前仓位状态并执行相应操作 if self.turtle.TrigerTime == 0: # 空仓状态 if self.turtle.system1EnterNormal( self.turtle.PriceNow, self.turtle.Donchian_20_up, self.turtle.BreakOutLog ): self.Buy_stock(self.turtle.PriceNow) elif self.turtle.system1EnterSafe( self.turtle.PriceNow, self.turtle.Donchian_50_up ): self.Buy_stock(self.turtle.PriceNow) elif 1 <= self.turtle.TrigerTime <= 3: # 加仓状态 if self.turtle.system1EnterNormal( self.turtle.PriceNow, self.turtle.Donchian_20_up, self.turtle.BreakOutLog ): self.Buy_stock(self.turtle.PriceNow) elif self.turtle.system1EnterSafe( self.turtle.PriceNow, self.turtle.Donchian_50_up ): self.Buy_stock(self.turtle.PriceNow) elif self.turtle.add(self.turtle.PriceNow): self.Buy_stock(self.turtle.PriceNow) elif self.turtle.system_1_stop(self.turtle.PriceNow): self.stop_sale_stock(self.turtle.PriceNow) elif self.turtle.system_1_Out( self.turtle.PriceNow, self.turtle.Donchian_10_down ): self.out_sale_stock(self.turtle.PriceNow) elif self.turtle.TrigerTime == 4: # 满仓状态 if self.turtle.system_1_stop(self.turtle.PriceNow): self.stop_sale_stock(self.turtle.PriceNow) elif self.turtle.system_1_Out( self.turtle.PriceNow, self.turtle.Donchian_10_down ): self.out_sale_stock(self.turtle.PriceNow) # 等待 1 分钟后下一次循环 time.sleep(60) def Start_short_system(self): """启动short系统 """ # ------------------准备阶段-------------------- # 获取数据或读取数据 -- 计算ATR Donchian 20 50 up, 20 down self.turtle.get_ready(100) self.turtle.N = self.turtle.CurrentData['ATR'].iloc[-1] self.turtle.Donchian_20_up = self.turtle.CurrentData['Donchian_20_upper'].iloc[-1] self.turtle.Donchian_50_up = self.turtle.CurrentData['Donchian_50_upper'].iloc[-1] self.turtle.Donchian_10_down = self.turtle.CurrentData['Donchian_10_lower'].iloc[-1] self.turtle.CalPositionSize() # ------------------实时监测阶段-------------------- # 9:00 1、判断是否是新的一周，是则重新计算Position Size # 判断是否是新的一周 if datetime.now().weekday() == 0: self.turtle.CalPositionSize() # 每分钟获取一次数据，判断是否触发条件 9:30-11:30 13:00-15:00 self.run_short_trading_loop() # ------------------结束阶段-------------------- # 数据库更新当天数据，增加ATR、donchian数据 pass time.sleep(16.5*600) if __name__ == '__main__': user_email = "guoyize2209@163.com" t = TurtleTrading('513300', "etf", 0.25, 100000, 200000) # t.get_ready(100) a = TurtleTrading_OnTime(t) a.Start_S1_system() # # 全是股票 # stock_zh_a_spot_df = ak.stock_zh_a_spot_em() # # stock_zh_a_spot_df.to_csv("stock_zh_a_spot.txt", sep="\t", index=False, encoding="utf-8") # stock_zh_a_spot_df = stock_zh_a_spot_df.dropna(subset=['最新价']) # print(stock_zh_a_spot_df) # # 全是基金 # etf_data = ak.fund_etf_spot_em() # etf_data = etf_data.dropna(subset=['最新价']) # etf_data.to_csv("fund_etf_spot.txt", sep="\t", index=False, encoding="utf-8") # print(etf_data)