以下是使用 Python 结合 `pandas`、`numpy`、`matplotlib`、`tushare`、`seaborn`、`random`、`datetime` 库完成你所要求任务的代码： ```python import tushare as ts import pandas as pd import numpy as np import matplotlib.pyplot as plt import seaborn as sns import random from datetime import datetime # 设置 tushare token ts.set_token('your_token') pro = ts.pro_api() # 获取沪深 300 指数成分股 hs300 = pro.index_weight(index_code='000300.SH', start_date='20230101', end_date='20240101') stock_list = hs300['con_code'].tolist() # 随机选择 4 只股票 random_stocks = random.sample(stock_list, 4) # 获取股票每日收盘价格数据 start_date = '20230101' end_date = '20240101' dfs = [] for stock in random_stocks: df = pro.daily(ts_code=stock, start_date=start_date, end_date=end_date) df = df.sort_values(by='trade_date') df.set_index('trade_date', inplace=True) dfs.append(df[['close']].rename(columns={'close': stock})) # 合并收盘价数据 combined_df = pd.concat(dfs, axis=1) # 输出股票名称与代码 stock_info = pro.stock_basic(exchange='', list_status='L', fields='ts_code,name') for stock in random_stocks: name = stock_info[stock_info['ts_code'] == stock]['name'].values[0] print(f"股票代码: {stock}, 股票名称: {name}") # 以初始股价 200 画出可比价格走势图 normalized_df = combined_df / combined_df.iloc[0] * 200 plt.figure(figsize=(12, 6)) plt.plot(normalized_df) plt.title('可比价格走势图') plt.xlabel('日期') plt.ylabel('价格') plt.legend(random_stocks) plt.show() # 计算每支股票收益与风险 returns = combined_df.pct_change().dropna() mean_returns = returns.mean() cov_matrix = returns.cov() # 求解投资组合最优组合 num_portfolios = 10000 weights_record = [] returns_record = [] volatility_record = [] for _ in range(num_portfolios): weights = np.random.random(len(random_stocks)) weights /= np.sum(weights) portfolio_return = np.dot(weights, mean_returns) portfolio_volatility = np.sqrt(np.dot(weights.T, np.dot(cov_matrix, weights))) returns_record.append(portfolio_return) volatility_record.append(portfolio_volatility) weights_record.append(weights) portfolios = pd.DataFrame({ '收益率': returns_record, '波动率': volatility_record, '权重': weights_record }) # 找到夏普比率最大的组合 risk_free_rate = 0.02 portfolios['夏普比率'] = (portfolios['收益率'] - risk_free_rate) / portfolios['波动率'] optimal_portfolio = portfolios.loc[portfolios['夏普比率'].idxmax()] print("最优组合权重:", optimal_portfolio['权重']) # 画出投资组合有效前沿 plt.figure(figsize=(12, 6)) plt.scatter(portfolios['波动率'], portfolios['收益率'], c=portfolios['夏普比率'], cmap='viridis') plt.colorbar(label='夏普比率') plt.scatter(optimal_portfolio['波动率'], optimal_portfolio['收益率'], color='red', marker='*', s=200, label='最优组合') plt.title('投资组合有效前沿') plt.xlabel('波动率') plt.ylabel('收益率') plt.legend() plt.show() # 拓展随机选 5 - 10 只票画投资组合有效前沿 for num_stocks in range(5, 11): random_stocks = random.sample(stock_list, num_stocks) dfs = [] for stock in random_stocks: df = pro.daily(ts_code=stock, start_date=start_date, end_date=end_date) df = df.sort_values(by='trade_date') df.set_index('trade_date', inplace=True) dfs.append(df[['close']].rename(columns={'close': stock})) combined_df = pd.concat(dfs, axis=1) returns = combined_df.pct_change().dropna() mean_returns = returns.mean() cov_matrix = returns.cov() num_portfolios = 10000 weights_record = [] returns_record = [] volatility_record = [] for _ in range(num_portfolios): weights = np.random.random(len(random_stocks)) weights /= np.sum(weights) portfolio_return = np.dot(weights, mean_returns) portfolio_volatility = np.sqrt(np.dot(weights.T, np.dot(cov_matrix, weights))) returns_record.append(portfolio_return) volatility_record.append(portfolio_volatility) weights_record.append(weights) portfolios = pd.DataFrame({ '收益率': returns_record, '波动率': volatility_record, '权重': weights_record }) risk_free_rate = 0.02 portfolios['夏普比率'] = (portfolios['收益率'] - risk_free_rate) / portfolios['波动率'] optimal_portfolio = portfolios.loc[portfolios['夏普比率'].idxmax()] plt.figure(figsize=(12, 6)) plt.scatter(portfolios['波动率'], portfolios['收益率'], c=portfolios['夏普比率'], cmap='viridis') plt.colorbar(label='夏普比率') plt.scatter(optimal_portfolio['波动率'], optimal_portfolio['收益率'], color='red', marker='*', s=200, label='最优组合') plt.title(f'{num_stocks} 只股票投资组合有效前沿') plt.xlabel('波动率') plt.ylabel('收益率') plt.legend() plt.show() ``` ### 代码说明： 1. **获取沪深 300 指数成分股**：使用 `tushare` 库获取沪深 300 指数成分股列表。 2. **随机选择股票**：从成分股列表中随机选择 4 只股票。 3. **获取收盘价格数据**：使用 `tushare` 库获取每只股票的每日收盘价格数据，并合并数据。 4. **画出可比价格走势图**：将每只股票的价格归一化到初始股价 200，然后画出走势图。 5. **计算收益与风险**：计算每只股票的每日收益率和协方差矩阵。 6. **求解最优组合**：通过随机生成大量投资组合，计算每个组合的收益率、波动率和夏普比率，找到夏普比率最大的组合。 7. **画出有效前沿**：将所有投资组合的收益率和波动率绘制成散点图，标记出最优组合。 8. **拓展到 5 - 10 只股票**：重复上述步骤，随机选择 5 - 10 只股票，画出投资组合有效前沿。 ###