What Is Spearman rank correlation?
Spearman rank correlation, also called Spearman’s rho, is a non-parametric measure of the monotonic relationship between two variables. It is calculated as the Pearson correlation of the variables’ ranks rather than of their raw values. Its value ranges from -1 to 1: +1 means the observations have exactly the same ordering, -1 means exactly the reverse ordering, and 0 indicates no monotonic association.
Unlike ordinary (Pearson) correlation, Spearman rank correlation does not require a linear relationship and is less affected by extreme values. It measures whether one variable generally rises as the other rises, even if the size of each change is not proportional.
In quantitative finance, it is commonly used to calculate rank IC: at a rebalance date, rank the assets by a signal and by their forward returns, then correlate the two sets of ranks. This makes it useful for cross-sectional factor research and machine-learning stock selection, where correctly ordering assets can matter more than predicting exact returns.
See also
Research and literature
Charles Spearman, The Proof and Measurement of Association between Two Things, The American Journal of Psychology 15(1), 1904.