Direct Answer
The Pearson correlation coefficient measures the strength and direction of the linear relationship between two variables, scaled to lie between -1 and +1. It is covariance divided by the product of the two standard deviations. +1 is a perfect increasing linear relation, -1 a perfect decreasing one and 0 no linear relation.
Correlation Coefficient: Formula, Meaning, Signals, Examples and How to Use It
How Is Correlation Coefficient Calculated?
The Pearson correlation coefficient measures the strength and direction of the linear relationship between two variables, scaled to lie between -1 and +1. It is covariance divided by the product of the two standard deviations.
Formula: r = (N*sum(xy) - sum(x)*sum(y)) / sqrt( (N*sum(x^2) - sum(x)^2) * (N*sum(y^2) - sum(y)^2) )
Inputs:
- Two aligned series
- Number of observations N
Worked example. For x values 1 to 5 and y values 2, 4, 5, 4, 5 the correlation is 0.775. The inputs are illustrative numbers, not market data.
Reading it. +1 is a perfect increasing linear relation, -1 a perfect decreasing one and 0 no linear relation. Squaring it gives the share of variance in one variable explained by a linear fit on the other.
Limits. It misses nonlinear relationships, is affected by outliers, and says nothing about causation. A single full-sample value hides changes over time.
What is Correlation Coefficient?
The Pearson correlation coefficient measures the strength and direction of the linear relationship between two variables, scaled to lie between -1 and +1. It is covariance divided by the product of the two standard deviations.
How to use Correlation Coefficient
+1 is a perfect increasing linear relation, -1 a perfect decreasing one and 0 no linear relation. Squaring it gives the share of variance in one variable explained by a linear fit on the other. A reading alone is never a buy or sell signal; it needs context from price structure, volume, volatility and the broader market regime.
Strengths
- Dividing covariance by both standard deviations bounds the result between -1 and +1, so the scale is fixed.
- Squaring it gives the share of variance a simple linear fit explains, which links it to R-squared.
- Reproducible when the formula and data convention are fixed.
Weaknesses and limitations
- Two series can show a correlation of zero and still be perfectly related, for instance when one is the square of the other.
- The measure weights all points equally and does not distinguish the co-movement of large moves from that of small ones.
- Correlated indicators can create false confidence by repeating the same underlying information.
Editorial and risk note
This page is educational content, not individualized investment advice. It should not imply guaranteed prediction, accuracy or outperformance.
Frequently Asked Questions
Is Correlation Coefficient a buy or sell signal?
No. +1 is a perfect increasing linear relation, -1 a perfect decreasing one and 0 no linear relation. It misses nonlinear relationships, is affected by outliers, and says nothing about causation. A trade built on it still needs a trigger, an invalidation rule, a position size and realistic execution assumptions.
What is the best setting for Correlation Coefficient?
Correlation Coefficient is computed from two aligned series; number of observations N, so the choices that matter are the data convention and the window it is read over. Test those across instruments, regimes and out-of-sample periods instead of fitting one historical winner.
Can Correlation Coefficient be used by itself?
It misses nonlinear relationships, is affected by outliers, and says nothing about causation. A single full-sample value hides changes over time. For that reason it is read next to measures of direction, participation, valuation, liquidity and risk, not alone.
Does Correlation Coefficient work on every timeframe?
The calculation (r = (N*sum(xy) - sum(x)*sum(y)) / sqrt( (N*sum(x^2) - sum(x)^2) * (N*sum(y^2) - sum(y)^2) )) can be run on any bar size, but behavior is not portable: a reading on a five-minute chart describes a very different market window from the same reading on a daily chart. Validate the exact timeframe before relying on it.
Why does Correlation Coefficient give false signals?
Two series can show a correlation of zero and still be perfectly related, for instance when one is the square of the other. The measure weights all points equally and does not distinguish the co-movement of large moves from that of small ones. Lag, noise and regime changes add to this, because market participants react to new information after the reading is calculated.
How should Correlation Coefficient be backtested?
Rebuild Correlation Coefficient from two aligned series using point-in-time data, write the rules before testing, include delisted securities where relevant, model realistic fills and costs, reserve a validation sample and compare with a simpler baseline.