Can someone explain correlation vs. causation with a finance example? My professor keeps emphasizing this.

This is a foundational concept that the CFA exam loves to test in tricky ways. Let's break it down.

What Correlation Tells You

The correlation coefficient r measures the linear association between two variables. It ranges from −1 to +1. An r = 0.72 between GDP growth and stock returns tells you they tend to move together, but it says nothing about whether one causes the other.

Three Reasons Correlation ≠ Causation

1. Reverse Causality: Maybe stock markets drive GDP (wealth effect) rather than GDP driving stocks.
2. Omitted Variable (Confounding): Both GDP and stocks might be driven by a third factor — say, expansionary monetary policy.
3. Spurious Correlation: Unrelated variables can show correlation by coincidence, especially with small samples or trending data.

Finance Example:

Sylvan Analytics finds a correlation of r = 0.65 between a company's advertising spend and revenue growth over 40 quarters. The analyst concludes 'increasing advertising by $1M will boost revenue by $3.2M.' This is a causal claim from correlational data — a classic error.

Possible confounders:

• Both ad spend and revenue rise during economic expansions
• Companies increase ads after good revenue (reverse causality)
• A new CEO drives both strategy changes simultaneously

What r² Tells You

The coefficient of determination r² = 0.72² = 0.5184, meaning about 52% of the variation in stock returns is associated with (not caused by) variation in GDP growth.

CFA Exam Traps:

• Statements like 'X causes Y because they are highly correlated' — always wrong
• Confusing statistical significance with economic significance
• Ignoring non-linear relationships (r only captures linear association)

For deeper practice on regression and correlation, check out our CFA Level I question bank.