Statistics & Backtest Pitfalls
Beta, Sharpe, portfolio vol, correlation bounds and the biases that kill a backtest the quant-research screen in full.
The scenario
Quant research round. You'll be handed summary statistics and a strategy's numbers and asked to reason about them cold compute the risk metrics, then explain why a shiny backtest might be a mirage. This is where lookahead, survivorship and multiple-testing separate researchers from over-fitters.
Where this shows up
A rapid statistics quiz plus backtest-pitfall reasoning is the standard quant-research first round G-Research's 10-question quiz and Citadel's timed multiple-choice are the canonical examples.
Firms such as G-Research, Two Sigma, Citadel.
DeskPrep is not affiliated with, endorsed by, or sponsored by any named firm. Firm names are used for illustrative, educational purposes only and do not imply that these materials are official assessments of, or are connected with, those firms.
The drill
Compute each figure (use population variance where variance is asked; annualise with √252). For the concept questions, pick the single best answer or select all that apply. Numeric answers are graded to a small tolerance.
Sample question
Compute the population variance (divide by n) of the three numbers 4, 8, 6.
Included with the Quant pack
The full graded drill is part of the Quant Research & Quant Trading pack a one-time purchase unlocks every concept, coding guide, interview playbook and take-home in this track.
already purchased? sign in →