diff --git a/language_learning/python/2019.10.27_stock_date_from_tushare/stock_date_from_tushare.py b/language_learning/python/2019.10.27_stock_date_from_tushare/stock_date_from_tushare.py deleted file mode 100644 index c22efca..0000000 --- a/language_learning/python/2019.10.27_stock_date_from_tushare/stock_date_from_tushare.py +++ /dev/null @@ -1,106 +0,0 @@ -""" -This code is supported by the website: https://www.guanjihuan.com -The newest version of this code is on the web page: https://www.guanjihuan.com/archives/706 -""" - -import numpy as np -import time -import matplotlib.pyplot as plt -import tushare as ts - - -def main(): - start_clock = time.perf_counter() - pro = ts.pro_api('到官网上注册，寻找Token填在这里!') - print('\n我的策略：见好就收，遇低抄底。\n' - ' 【卖出】买入后涨了5%就卖出\n' - ' 【买入】卖出后跌了5%就买入\n' - '注：第一天必须买进，最后一天前必须卖出(为了与不操作的做对比)\n') - number = 1 - for i in range(number): - data = pro.stock_basic(exchange='', list_status='L', fields='ts_code,symbol,name,area,industry,list_date') # 所有股票列表 - # print(data.columns) # 查看该数据的表头 - # print(data) # 3688多行的股票数据 - i = 1 # 查看第二行数据“万科A”股 - ts_code = data.values[i, 0] # 股票代码 - stock = data.values[i, 2] # 股票名称 - industry = data.values[i, 4] # 属于哪个行业 - start_date = '20110101' # 开始时间 - end_date = '20191027' # 结束时间 - df = pro.daily(ts_code=ts_code, start_date=start_date, end_date=end_date) # 查看该股票的日线数据 - # print(df.columns) # 查看该数据的表头 - # print(df) # 查看该股票的日线数据 - close = np.array(list(reversed(df.values[:, 5]))) # 提取出收盘价，并按时间顺序排列，从过去到现在 - pct_chg = np.array(list(reversed(df.values[:, 8]))) # 提取出涨跌幅，并按时间顺序排列，从过去到现在 - # print(df.columns[5], '=', close, '\n') # 查看收盘价 - # print(df.columns[8], '=', pct_chg, '\n') # 查看涨跌幅 - profit, profit_no_operation, times, invest_money, buy_time_all, sell_time_all = back_test(close.shape[0], close, pct_chg) - # 调用回测函数，返回了“利润，未操作的利润， 按该策略操作了几次， 总投资金额， 按该策略买的时间， 按该策略卖的时间”的值 - print('\n------股票：', stock, ts_code, industry, '[买入市值=%7.2f' % invest_money, ']------') - print('回测时间段：', start_date, '-', end_date) - print('操作后利润= %6.2f' % profit, ' 买入（卖出）次数=', times, ' ') - print('不操作利润= %6.2f' % profit_no_operation, '（第一天买入，最后一天卖出，中间未操作）') - end_clock = time.perf_counter() - print('CPU执行时间=', end_clock - start_clock, 's') - plt.figure(1) - plt.title('Stock Code: '+ts_code+' (red point: buy, green point: sell)') - plt.grid() - plt.plot(range(close.shape[0]), close, '-') - for i in buy_time_all: - plt.plot(i, close[int(i)], 'or', markersize=13) # 红色是买进的点 - for i in sell_time_all: - plt.plot(i, close[int(i)], 'dg', markersize=13) # 绿色是卖出的点 - plt.show() - - -def back_test(days, close, pct_chg, money_in=10000): # 定义该策略的回测效果（按旧数据检查该策略是否有效） - money_in_amount = int(money_in/close[0]) # 投资金额换算成股票股数 - invest_money = close[0]*money_in_amount # 实际买了股票的金额 - profit_no_operation = (close[close.shape[0]-1]-close[0])*money_in_amount # 不操作的利润 - position = -1 # 买入还是卖出的状态，默认卖出 - total_profit = 0 - times = 0 - current_buy_pct = -999 - current_sell_pct = 999 - buy_time_all = np.array([]) - sell_time_all = np.array([]) - for i in range(days): # 总天数 - if i == 0: # 第一天，满仓买买买！为了和不操作的对比，第一天就要买入。 - buy_time = i # 买入时间 - buy_time_all = np.append(buy_time_all, [buy_time], axis=0) # 买入时间存档 - position = 1 # 标记为买入状态 - print('------------------第', buy_time, '天买进-------------') - else: - profit = 0 - if position == 1: # 买入状态 - current_buy_pct = (close[i]-close[buy_time])/close[buy_time]*100 # 买入后的涨跌情况 - # print('当前买进后的涨跌情况：第', i, '天=', current_buy_pct) - if position == 0: # 卖出状态 - current_sell_pct = (close[i]-close[sell_time])/close[sell_time]*100 # 卖出后的涨跌情况 - - if current_sell_pct < -5 and position == 0: # 卖出状态，且卖出后跌了有3%，这时候买入 - buy_time = i # 买入时间 - buy_time_all = np.append(buy_time_all, [buy_time], axis=0) # 买入时间存档 - print('------------------第', buy_time, '天买进-------------') - position = 1 # 标记为买入状态 - continue - - if current_buy_pct > 5 and position == 1: # 买入状态，且买入后涨了有3%，这时候卖出 - sell_time = i # 卖出时间 - sell_time_all = np.append(sell_time_all, [sell_time], axis=0) # 卖出时间存档 - print('----------第', sell_time, '天卖出，持有天数：', sell_time-buy_time, '--------------\n') - position = 0 # 标记为卖出状态 - profit = close[sell_time]-close[buy_time] # 赚取利润 - times = times + 1 # 买入（卖出）次数加1 - total_profit = total_profit + profit*money_in_amount # 计算总利润 - if position == 1: # 最后一天如果是买入状态，则卖出 - profit = close[i]-close[buy_time] # 赚取利润 - total_profit = total_profit + profit # 计算总利润 - times = times + 1 # 买入（卖出）次数加1 - print('--------------第', i, '天（最后一天）卖出，持有天数：', sell_time-buy_time, '--------------\n') - sell_time_all = np.append(sell_time_all, [i], axis=0) # 卖出时间存档 - return total_profit, profit_no_operation, times, invest_money, buy_time_all, sell_time_all - - -if __name__ == '__main__': - main()