The go-to non-parametric test for comparing two independent groups when your data isn't normally distributed.
The Mann-Whitney U Test is the non-parametric version of the independent T-Test. Instead of comparing means, it compares the **ranks** of the data from two groups. Think of it as lining up all data points from both groups and checking if one group's values are consistently ranked higher than the other's.
Use this test when you want to compare two independent groups but your data **does not follow a normal distribution**. This is common with financial data like trade returns, which are often skewed and have "fat tails" (more extreme outcomes than a normal distribution would suggest).
This test is perfect for comparing groups where the data is skewed. A classic example is trade profitability, where you might have many small gains and a few very large gains, creating a long tail.
Example: A quant firm develops a new trading algorithm ('Algo B') and wants to see if it generates significantly different profits than their old one ('Algo A'). The profit distributions are known to be skewed (not normal). They use the Mann-Whitney U Test to determine if there's a statistical difference in the distribution of profits between the two algorithms.