Shows systematic relationship between two random variables. Values range from -∞ to +∞, indicating positive or negative relationship. Positive covariance shows variables moving in same direction. Negative covariance shows variables moving in opposite direction. Used in Cholesky decomposition and principal component analysis
Joint distribution shows all possible pairs of outputs of two random variables. Can be expressed as multivariate CDF or probability density/mass functions. Marginal distributions encode individual variable outputs
Measures linear relationship between variables. Returns values between -1 and 1. Uses mean and standard deviation, assumes Gaussian distribution. Sensitive to outliers and may lead to incorrect conclusions
Scatter plot visually displays relationships between variables using dots. Values are represented by dots positioned on Cartesian coordinates. Also known as scattergram, scatter graph, or scatter chart
xcorr(x,y) returns cross-correlation of two discrete-time sequences. xcorr(x) computes autocorrelation of a single vector. Function appends zeros to shorter vectors when lengths differ
Both measure linear relationships between random variables. Covariance shows direction of variable variation. Correlation measures both direction and strength of relationship