# Variance Reduction

*Quant & Pricing — Finicade finance glossary*

Variance reduction is the set of tricks that make Monte Carlo converge with far fewer paths. Antithetic variates pair each random draw with its mirror image; control variates exploit a related instrument with a known analytic price; low-discrepancy sequences fill the space more evenly than random numbers. Since Monte Carlo error falls only with the square root of paths, a technique that halves variance is worth four times the hardware.

**Also known as:** antithetic variates, control variates, quasi-Monte Carlo

**Related terms:** [Monte Carlo Simulation](https://finicade.com/glossary/monte-carlo), [Standard Error](https://finicade.com/glossary/standard-error), [Least Squares Monte Carlo](https://finicade.com/glossary/least-squares-monte-carlo), [Sampling](https://finicade.com/glossary/sampling), [Confidence Interval](https://finicade.com/glossary/confidence-interval)

**Taught in:** Quant Quest — Monte Carlo Simulation

Source: https://finicade.com/glossary/variance-reduction