Interview questions
1,322 questions in the style trading and research firms ask, from probability and brainteasers to options, statistics and coding. Every question is open to read, with its hints. The 50 in the free sample can be answered and marked on the spot, no account needed; open 3 of their worked solutions to try, then all of them with a free account.
496 questions · page 9 of 13 · Clear filters
Random question- 321The area under a Brownian path, I = ∫₀³Wₛ ds, has mean zero. What is its variance?
AdvancedStochastic calculusJane StreetCitadelTwo Sigma
- 322In the CIR model dr = κ(θ−r) dt+σ√r dW with θ = 0.04 and σ = 0.3, what is the…
AdvancedStochastic calculusJane StreetCitadelTwo Sigma
- 323Under ℙ, W is standard Brownian motion.
AdvancedStochastic calculusJane StreetCitadelTwo Sigma
- 324What does the Feynman–Kac theorem connect?
AdvancedStochastic calculusJane StreetCitadelTwo Sigma
- 325Why price a bond option under the forward measure rather than the risk-neutral one?
ExpertStochastic calculusJane StreetCitadelCitadel Securities
- 326You deepen your trees from depth 3 to depth 12. What happens to bias and variance?
FoundationMachine learningCitadelTwo Sigma
- 327A node holds 40 positive and 60 negative examples. What is its Gini impurity?
FoundationMachine learningJane StreetCitadelD. E. Shaw
- 328A model’s predictions at a point have bias 0.3 and variance 0.2, and the…
FoundationMachine learningCitadelTwo Sigma
- 329As you train longer, training error keeps falling while validation error starts rising.
FoundationMachine learningCitadelTwo Sigma
- 330A node holds 40 positive and 60 negative examples.
FoundationMachine learningCitadelTwo Sigma
- 331A logistic regression has weights w = (2,−1) and bias −1.
FoundationMachine learningCitadelTwo Sigma
- 332One-dimensional points 1,2,3,10,11,12 are clustered by k-means with k = 2 and…
FoundationMachine learningCitadelTwo Sigma
- 333A point’s mean distance to others in its own cluster is 2, and to the nearest…
FoundationMachine learningCitadelTwo Sigma
- 334A classifier has 30 true positives, 20 false positives and 10 false negatives.
FoundationMachine learningCitadelTwo Sigma
- 335A classifier predicts probability 0.8 for an example that is in fact positive.
FoundationMachine learningCitadelTwo Sigma
- 336How many learnable parameters does a fully connected layer from 100 inputs to…
FoundationMachine learningCitadelTwo Sigma
- 337A softmax output layer receives logits (2,1,0).
FoundationMachine learningCitadelTwo Sigma
- 338Why is ordinary shuffled k-fold cross-validation wrong on a financial time series?
AppliedMachine learningCitadelTwo SigmaPoint72 and CubistFree solution
- 339Why do gradient-boosted trees usually beat deep networks on daily…
AppliedMachine learningCitadelTwo SigmaD. E. Shaw
- 340In a ten-dimensional unit cube, what fraction of the volume lies within 0.1 of some face?
AppliedMachine learningJane StreetCitadelTwo Sigma
- 341A support vector machine has |w|= 4. What is the width of its margin?
AppliedMachine learningJane StreetCitadelD. E. Shaw
- 342Gradient descent with momentum uses a learning rate of 0.01 and β = 0.9.
AppliedMachine learningJane StreetCitadelHudson River Trading
- 343A batch normalisation layer follows a fully connected layer of 128 units.
AppliedMachine learningJane StreetCitadelJump Trading
- 344A classifier flags 500 cases, of which 40 are true positives, and misses 60 positives.
AppliedMachine learningJane StreetCitadelD. E. Shaw
- 345Data are uniform in the unit hypercube in 10 dimensions.
AppliedMachine learningJane StreetCitadelTwo Sigma
- 346A node has 50 positives and 50 negatives.
AppliedMachine learningCitadelTwo Sigma
- 347In gradient boosting, what happens if you halve the learning rate (shrinkage)…
AppliedMachine learningCitadelTwo Sigma
- 348Twenty percent of messages are spam.
AppliedMachine learningCitadelTwo Sigma
- 349What does the EM algorithm guarantee when fitting a Gaussian mixture model?
AppliedMachine learningCitadelTwo Sigma
- 350An anomaly detector has sensitivity 90% and specificity 95%.
AppliedMachine learningJane StreetCitadelTwo Sigma
- 351You apply a strictly increasing transformation to a classifier’s scores.
AppliedMachine learningCitadelTwo Sigma
- 352Gradient descent on f(x) = x² starts at x = 1 with learning rate 0.1.
AppliedMachine learningCitadelTwo Sigma
- 353A quadratic loss has Hessian eigenvalues 4 and 0.5.
AppliedMachine learningCitadelTwo Sigma
- 354A convolutional layer uses 3 × 3 kernels, 16 input channels and 32 output…
AppliedMachine learningCitadelTwo Sigma
- 355A regularised linear model beats your tuned gradient-boosting ensemble out of sample.
AdvancedMachine learningCitadelTwo SigmaAQR
- 356Two features are nearly identical, and a random forest reports low importance for both.
AdvancedMachine learningCitadelTwo SigmaAQR
- 357What does the kernel trick actually buy you?
AdvancedMachine learningJane StreetCitadelJump Trading
- 358You increase the batch size from 32 to 512 without changing anything else. What happens?
AdvancedMachine learningJane StreetCitadelTwo Sigma
- 359Why did rectified linear units largely replace sigmoids in deep networks?
AdvancedMachine learningJane StreetCitadelD. E. Shaw
- 360You cluster a universe of assets from their correlation matrix.
AdvancedMachine learningJane StreetCitadelTwo Sigma