Python Data Analysis Cookbook

Correlating individual stocks with the broader market

When we define a stock market or index, we usually choose stocks that are similar in some way. For instance, the stocks might be in the same country or continent. The position of birds can be roughly estimated from the position of the flock they belong to. Similarly, we expect stock returns to be correlated to their market, although not necessarily perfectly.

We will explore the following metrics:

• The most obvious metric is purportedly the correlation coefficient of the individual stock returns and the S&P 500 index.
• Another metric is the slope obtained from linear regression instead of correlation.
• We can also analyze squared differences of returns somewhat similar to squared errors in regression ...