[Python] How to Compute Pairwise Correlations for Thousands of Stocks in Batch

I would like to compute the pairwise correlation of stock return data using DolphinDB. There are thousands of stocks. How can I achieve this? I have already implemented a version in Python:

import pandas as pd
import numpy as np

# suppose you have a table like this：
# AAPL MSFT GOOGL ...
# 2021-01-01 0.01 0.02 -0.01
# 2021-01-02 0.00 0.01 0.02
# ...

tickers = ['AAPL', 'MSFT', 'GOOGL', 'AMZN', 'META']
dates = pd.date_range(start='2021-01-01', periods=100)
np.random.seed(0)
data = np.random.randn(100, len(tickers)) / 100 # daily ratios

df = pd.DataFrame(data, index=dates, columns=tickers)

correlation_matrix = df.corr()


print(correlation_matrix.head())

Continue reading...