The XTX Markets Onsite Probability, Statistics & Linear Algebra interview is stage 6 of 6 in the XTX Markets interview process, after the Onsite Machine Learning / Research Deep-Dive. Format: 60 min · whiteboard. A super-day round on XTX's mathematical foundations: derive from first principles, explain why estimators behave as they do, and move between geometry, algebra and, when asked, a few lines of code. - Probability and statistics: MLE and common distributional families, sampling and estimator properties, why we divide by n−1, hypothesis testing and multiple-testing corrections. - Linear algebra: covariance and positive-semidefinite matrices, the SVD and best low-rank approximation, eigenstructure, QR or Gram-Schmidt orthogonalization. - Set up the objective or probability space explicitly, turn the crank, then sanity-check with limits, symmetry and degenerate cases: catching your own error beats a lucky final number.
Preparing for this round? Open the XTX Markets Onsite Probability, Statistics & Linear Algebra in the interview funnel. A free account saves your progress.
What the Onsite Probability, Statistics & Linear Algebra tests
A super-day round on XTX's mathematical foundations: derive from first principles, explain why estimators behave as they do, and move between geometry, algebra and, when asked, a few lines of code.
- Probability and statistics: MLE and common distributional families, sampling and estimator properties, why we divide by n−1, hypothesis testing and multiple-testing corrections.
- Linear algebra: covariance and positive-semidefinite matrices, the SVD and best low-rank approximation, eigenstructure, QR or Gram-Schmidt orthogonalization.
- Set up the objective or probability space explicitly, turn the crank, then sanity-check with limits, symmetry and degenerate cases: catching your own error beats a lucky final number.
How to prepare for the Onsite Probability, Statistics & Linear Algebra
- Derive estimators from scratch. MLE for the common families, why we divide by n−1, the OLS estimator — set up the objective and turn the crank.
- Matrix structure is a recurring theme. PSD covariance, the SVD and best low-rank approximation, eigenvalues of structured (equicorrelation) matrices. Be fluent moving between geometry and algebra.
- Sanity-check with limits and symmetry. Catching your own error (n→∞, degenerate cases) scores better than a lucky final number.
Final-week checklist
- Drill speed on fundamentals (MLE, OLS, ridge/lasso, PCA/SVD, log-sum-exp) — the OA punishes hesitation and you reportedly can't go back to a question.
- Prepare one research / data project you can defend end to end, including how you validated it out-of-sample.
- Be ready to derive, not recall: normal equations, the first principal component, why L1 is sparse, why we divide by n−1.
Further reading: How quant superdays work · Behavioral questions at trading firms · Market-making game interview guide.
Prepare with QuantVault
The same round at similar firms
Compare with similar firms
Frequently asked questions
What is the XTX Markets Onsite Probability, Statistics & Linear Algebra interview?
It is stage 6 of 6 in the XTX Markets interview process, after the Onsite Machine Learning / Research Deep-Dive. Format: 60 min · whiteboard. A super-day round on XTX's mathematical foundations: derive from first principles, explain why estimators behave as they do, and move between geometry, algebra and, when asked, a few lines of code.
How long is the XTX Markets Onsite Probability, Statistics & Linear Algebra interview?
Candidates report: 60 min · whiteboard.
How should I prepare for the XTX Markets Onsite Probability, Statistics & Linear Algebra interview?
Derive estimators from scratch. MLE for the common families, why we divide by n−1, the OLS estimator — set up the objective and turn the crank. Matrix structure is a recurring theme. PSD covariance, the SVD and best low-rank approximation, eigenvalues of structured (equicorrelation) matrices. Be fluent moving between geometry and algebra.
What comes after the Onsite Probability, Statistics & Linear Algebra?
Nothing: it is the last of the 6 stages QuantVault maps for XTX Markets.
Practice the XTX Markets Onsite Probability, Statistics & Linear Algebra round
Open the round in the XTX Markets funnel: the round notes, prep, and every other stage of the process.