寻找美股500强每只股票与自身的相关性，如何防止重新计算相关性？

Question

这是较大部分代码的一部分，其中仅包括数据下载、数据操作和排序。我怎样才能让它只比较每只股票一次，例如，只比较微软与苹果，而不是微软与苹果以及苹果与微软？

index = 0
ticker_col = np.array([])
compare_col = np.array([])
correlation_col = np.array([])

for ticker in tickers:
    ticker = ticker[:-1]
    for compare in tickers:
        compare = compare[:-1]

        if ticker == compare:
            break
        index += 1

        compare_list = pd.read_csv('C:/stockpredictions/stocks_dfs/' + compare + '.csv')['Adj Close']
        ticker_list = pd.read_csv('C:/stockpredictions/stocks_dfs/' + ticker + '.csv')['Adj Close']

        #pearsons r correlation formula
        correlation = df['Data1'].corr(df['Data2'])

Answer 1

我添加了

enumerate

用于将当前索引存储到

i

变量中（您可以使用它来代替在第二个循环中使用的

index

变量）。 第二个循环现在从

i+1

索引开始，以防止值之间进行两次比较。

index = 0
ticker_col = np.array([])
compare_col = np.array([])
correlation_col = np.array([])

for i, ticker in enumerate(tickers):
    ticker = ticker[:-1]
    for compare in tickers[i+1:]:
        compare = compare[:-1]

        if ticker == compare:
            break
        index += 1

        compare_list = pd.read_csv('C:/stockpredictions/stocks_dfs/' + compare + '.csv')['Adj Close']
        ticker_list = pd.read_csv('C:/stockpredictions/stocks_dfs/' + ticker + '.csv')['Adj Close']

        #pearsons r correlation formula
        correlation = df['Data1'].corr(df['Data2'])

您可以使用以下代码进行测试：

tickers = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]

for i, ticker in enumerate(tickers):
    for compare in tickers[i+1:]:
        print(ticker, compare)