Correlation becomes actionable only when it’s measured: it determines whether multiple trades are independent bets or the same exposure repeated across symbols. Portfolio risk is not the sum of individual risks; pairwise correlations add interaction terms that can inflate or reduce total variance.
The article breaks risk into diversifiable (instrument-specific) and systematic (shared) components, then uses Pearson correlation and a covariance matrix to compute correlation-adjusted portfolio risk. Examples show identical 1% positions ranging from ~0.45% to ~1.95% combined risk depending only on correlation.
Two MT5 programs implement this: a script that reads open positions, pulls recent H1 data, builds the covariance matrix, and prints naive vs true risk plus a “hidden factor”; and a background service that continuously monitors, displays a compact pa...
👉 Read | AlgoBook | @mql5dev