用python编写秩和比算法

以下是用Python编写的秩和比算法的示例代码：

def ranksum_test(x, y):
    """
    Conducts the rank sum test (Mann-Whitney U test) on two samples.

    x: list or array of the first sample
    y: list or array of the second sample

    Returns the test statistic U and the p-value.
    """
    from scipy.stats import rankdata

    n1, n2 = len(x), len(y)

    # Combine the samples and rank them
    all_data = np.concatenate([x, y])
    ranks = rankdata(all_data)

    # Calculate the sum of the ranks for each sample
    r1 = np.sum(ranks[:n1])
    r2 = np.sum(ranks[n1:])

    # Calculate U and adjust for ties
    U = n1 * n2 + n1 * (n1 + 1) / 2 - r1
    U_adj = min(U, n1 * n2 - U)

    # Calculate the p-value using the normal approximation
    z = (U_adj - n1 * n2 / 2) / np.sqrt(n1 * n2 * (n1 + n2 + 1) / 12)
    p = 2 * (1 - norm.cdf(abs(z)))

    return U, p

这个函数使用了Python的SciPy库来进行秩和比检验。该函数将两个样本组合在一起，并对它们进行排序，然后计算每个样本的秩之和。接下来，它使用秩和公式计算U统计量，并根据秩的大小调整U值，以便对于任何可能存在的平局进行了调整。最后，该函数使用正态近似计算p值，并返回U值和p值。