以下是一个使用Python语言结合`pandas`、`numpy`、`matplotlib`、`tushare`、`seaborn`、`random`、`datetime`等库完成你所需求任务的代码示例： ```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的token ts.set_token('your_token_here') pro = ts.pro_api() # 获取沪深300成分股列表 hs300 = pro.index_weight(index_code='000300.SH', start_date='20200101', end_date=datetime.now().strftime('%Y%m%d')) stock_codes = hs300['con_code'].tolist() # 随机选择4只股票 random_stocks_4 = random.sample(stock_codes, 4) # 导出收盘价格数据 close_prices_4 = pd.DataFrame() for stock in random_stocks_4: df = pro.daily(ts_code=stock, start_date='20200101', end_date=datetime.now().strftime('%Y%m%d')) df['trade_date'] = pd.to_datetime(df['trade_date']) df = df.set_index('trade_date') close_prices_4[stock] = df['close'] # 合并收盘价数据 close_prices_4 = close_prices_4.sort_index() # 画出可比价格走势图 plt.figure(figsize=(12, 6)) for stock in random_stocks_4: plt.plot(close_prices_4.index, close_prices_4[stock], label=stock) plt.title('Comparable Closing Prices of 4 Randomly Selected Stocks') plt.xlabel('Date') plt.ylabel('Closing Price') plt.legend() plt.show() # 计算每支股票的收益与风险 returns_4 = close_prices_4.pct_change().dropna() mean_returns_4 = returns_4.mean() cov_matrix_4 = returns_4.cov() # 求解投资组合的最优组合 num_portfolios = 10000 risk_free_rate = 0.02 portfolios = [] for _ in range(num_portfolios): weights = np.random.random(4) weights /= np.sum(weights) portfolio_return = np.sum(weights * mean_returns_4) portfolio_std_dev = np.sqrt(np.dot(weights.T, np.dot(cov_matrix_4, weights))) sharpe_ratio = (portfolio_return - risk_free_rate) / portfolio_std_dev portfolios.append([portfolio_return, portfolio_std_dev, sharpe_ratio, weights]) portfolios = pd.DataFrame(portfolios, columns=['Return', 'Std Dev', 'Sharpe Ratio', 'Weights']) optimal_portfolio_4 = portfolios.loc[portfolios['Sharpe Ratio'].idxmax()] # 画出投资组合的有效前沿 plt.figure(figsize=(12, 6)) plt.scatter(portfolios['Std Dev'], portfolios['Return'], c=portfolios['Sharpe Ratio'], cmap='viridis') plt.colorbar(label='Sharpe Ratio') plt.scatter(optimal_portfolio_4['Std Dev'], optimal_portfolio_4['Return'], color='r', marker='*', s=200, label='Optimal Portfolio') plt.title('Efficient Frontier of 4 Randomly Selected Stocks') plt.xlabel('Standard Deviation') plt.ylabel('Expected Return') plt.legend() plt.show() # 随机选择5 - 10只股票 num_stocks = random.randint(5, 10) random_stocks_5_10 = random.sample(stock_codes, num_stocks) # 导出收盘价格数据 close_prices_5_10 = pd.DataFrame() for stock in random_stocks_5_10: df = pro.daily(ts_code=stock, start_date='20200101', end_date=datetime.now().strftime('%Y%m%d')) df['trade_date'] = pd.to_datetime(df['trade_date']) df = df.set_index('trade_date') close_prices_5_10[stock] = df['close'] # 合并收盘价数据 close_prices_5_10 = close_prices_5_10.sort_index() # 计算每支股票的收益与风险 returns_5_10 = close_prices_5_10.pct_change().dropna() mean_returns_5_10 = returns_5_10.mean() cov_matrix_5_10 = returns_5_10.cov() # 求解投资组合的最优组合 portfolios_5_10 = [] for _ in range(num_portfolios): weights = np.random.random(num_stocks) weights /= np.sum(weights) portfolio_return = np.sum(weights * mean_returns_5_10) portfolio_std_dev = np.sqrt(np.dot(weights.T, np.dot(cov_matrix_5_10, weights))) sharpe_ratio = (portfolio_return - risk_free_rate) / portfolio_std_dev portfolios_5_10.append([portfolio_return, portfolio_std_dev, sharpe_ratio, weights]) portfolios_5_10 = pd.DataFrame(portfolios_5_10, columns=['Return', 'Std Dev', 'Sharpe Ratio', 'Weights']) optimal_portfolio_5_10 = portfolios_5_10.loc[portfolios_5_10['Sharpe Ratio'].idxmax()] # 画出投资组合的有效前沿 plt.figure(figsize=(12, 6)) plt.scatter(portfolios_5_10['Std Dev'], portfolios_5_10['Return'], c=portfolios_5_10['Sharpe Ratio'], cmap='viridis') plt.colorbar(label='Sharpe Ratio') plt.scatter(optimal_portfolio_5_10['Std Dev'], optimal_portfolio_5_10['Return'], color='r', marker='*', s=200, label='Optimal Portfolio') plt.title('Efficient Frontier of 5 - 10 Randomly Selected Stocks') plt.xlabel('Standard Deviation') plt.ylabel('Expected Return') plt.legend() plt.show() ``` ### 代码说明： 1. **获取沪深300成分股列表**：使用`tushare`库获取沪深300成分股的列表。 2. **随机选择股票**：从沪深300成分股中随机选择4只和5 - 10只股票。 3. **导出收盘价格数据**：使用`tushare`库获取每只股票的收盘价格数据，并合并到一个`DataFrame`中。 4. **画出可比价格走势图**：使用`matplotlib`库画出4只股票的可比价格走势图。 5. **计算每支股票的收益与风险**：计算每只股票的收益率和协方差矩阵。 6. **求解投资组合的最优组合**：使用蒙特卡罗模拟方法生成多个投资组合，并计算每个组合的收益率、标准差和夏普比率，找到夏普比率最大的最优组合。 7. **画出投资组合的有效前沿**：使用`matplotlib`库画出投资组合的有效前沿，并标记出最优组合。 ###