GARCH Model Quant Interview Guide
GARCH model quant interview guide for conditional variance, volatility persistence, parameter intuition, examples, limitations, and validation.
Candidates preparing for volatility and risk modeling prompts.
GARCH models conditional variance
A GARCH model lets volatility depend on recent shocks and past volatility. The intuition is that risk can be persistent even when expected returns are hard to forecast.
Persistence is the key idea
If volatility was high recently, a GARCH-style model can keep the next-period variance estimate elevated. This captures clustering better than a constant-variance assumption.
Concrete example
A large return shock may increase the estimated conditional variance for upcoming periods. That could affect risk limits or forecast intervals, even if the return direction remains uncertain.
Validation still matters
A volatility model should be tested against realized outcomes and risk needs. Distribution assumptions, parameter stability, and regime changes can all weaken performance.
Common mistakes
Candidates often present GARCH as a forecasting machine. It is better framed as a conditional variance model with assumptions that need empirical validation.
Practice the pattern
Use the LeetQuidity curriculum and calibration to turn this topic into a focused practice plan.