Regression is a statistical measure that estimates the strength of relationship between a dependent variable and a series of other variables. It can be used to understand the relationships between variables. An example in finance would be understanding the relationship between commodity prices and the stocks of businesses dealing in those commodities.
There was originally a regression model built directly into pandas, but it has been moved out into the StatsModels library. This shows a pattern common in pandas. Often pandas has concepts built into it, but as they mature they are deemed to fit most effectively into other Python libraries. This is both good and bad. It is initially great to have it directly in pandas, but as you upgrade to new versions of pandas it can break your code!