Covariance and correlation — the math of diversification

Covariance: do they move together?

Volatility (A2-02) describes one return series in isolation. Portfolios live or die by a different question: when stock A has a bad day, what does stock B tend to do? Covariance is that question as a number:

$\text{Cov}(X, Y) = E[(X - \mu_X)(Y - \mu_Y)]$

Read the inside term day by day for two stocks. On a day when both are above their means, the product of deviations is positive × positive. Both below: negative × negative — positive again. One up, one down: the product is negative. Covariance averages these products:

• Positive covariance — they tend to be on the same side of their means together (two US banks).
• Near zero — no linear tendency either way (a tech stock and wheat futures, roughly).
• Negative — one tends to be up when the other is down (equities and long-dated Treasuries, in some regimes).

Note $\text{Cov}(X, X) = E[(X-\mu_X)^2] = \text{Var}(X)$: variance is covariance with yourself.

Correlation: covariance you can interpret

Covariance has awkward units (return × return) and its magnitude scales with each asset's volatility, so "Cov = 0.0002" is uninterpretable on its own. Normalize away the scales:

$\rho = \frac{\text{Cov}(X, Y)}{\sigma_X \, \sigma_Y}, \qquad -1 \le \rho \le +1$

The bounds are a theorem (Cauchy–Schwarz), not a convention. $\rho = +1$ means an exact increasing linear relationship, $-1$ an exact decreasing one, $0$ no linear relationship (a perfect U-shape can have $\rho = 0$ — correlation only sees lines). Typical daily-return correlations: two large-cap US stocks 0.3–0.6; a stock with its own sector ETF 0.7–0.9.

The sample versions replace expectations with averages, with the same $N-1$ Bessel correction as A2-02 — numpy's np.cov uses ddof=1 by default (unlike np.std, which defaults to ddof=0; a classic gotcha):

cov_xy = np.cov(x, y)[0, 1]                       # 2x2 matrix; off-diagonal is Cov(x, y)
corr_xy = cov_xy / (x.std(ddof=1) * y.std(ddof=1))

The payoff: portfolio variance

From A1-04, for portfolio weights $a$ and $b$:

$\text{Var}(aX + bY) = a^2\,\text{Var}(X) + b^2\,\text{Var}(Y) + 2ab\,\text{Cov}(X, Y)$

Everything about diversification hides in that last term. Work the pretest example properly: $\sigma_X = \sigma_Y = 20\%$, $a = b = 0.5$, so the first two terms give $0.25 \times 0.04 + 0.25 \times 0.04 = 0.02$, and $\text{Cov} = \rho \times 0.2 \times 0.2 = 0.04\rho$:

The lesson of the table: diversification is a correlation play, not a headcount play. At $\rho = +1$ you get nothing — you own one asset wearing two tickers. Real diversification benefit only appears as $\rho$ drops below 1, and it's why a portfolio of 30 highly correlated tech stocks is far less diversified than 5 genuinely unrelated return streams.

Two honest warnings

Correlation is not causation — and it isn't stable either. Beyond the classic inferential trap (A1-05's conditional-is-not-causal, in linear clothing), financial correlations are estimated from a window and drift across regimes, just like volatility in A2-02.

"Correlations go to 1 in a crash." The trading floor version is exaggerated but points at something real: measured correlations between risky assets rise sharply in crises — assets that looked pleasantly independent in calm markets fall together when everyone deleverages at once. (One technical caveat: correlation measured conditional on high-volatility periods is mechanically higher even if the true relationship never changed, so part of the effect is a measurement artifact — but the economic core is real.) The practical consequence is brutal: diversification measured in calm markets overstates the protection you'll have in the storm, which is precisely when you need it. Never size positions assuming calm-period correlations hold in a tail event.

Forward pointer: regression beta (A2-06) is $\text{Cov}(r_{\text{stock}}, r_{\text{mkt}}) / \text{Var}(r_{\text{mkt}})$ — covariance rescaled by the market's variance instead of by both standard deviations. Correlation and beta are the same object under two different normalizations.

Try it

Map it

On paper, draw a concept map connecting: variance, covariance, correlation, Var(aX+bY), diversification, crisis correlation, and beta. Draw a labelled arrow for each relationship (e.g., which normalization turns covariance into correlation? into beta? which term of the portfolio-variance formula does diversification act through?).