Skip to content
Get started

Resources · Cheat sheet

The quant interview cheat sheet.

The formulas and results that come up again and again, on one page. Memorise them so your working memory is free for the problem - then follow the links to understand why each one is true, because interviewers will ask.

Probability rules

Conditioning guide →
RuleFormula
ComplementP(Ac)=1−P(A)P(A^c) = 1 - P(A)
UnionP(A∪B)=P(A)+P(B)−P(A∩B)P(A \cup B) = P(A) + P(B) - P(A \cap B)
ConditionalP(A∣B)=P(A∩B)P(B)P(A \mid B) = \dfrac{P(A \cap B)}{P(B)}
Total probabilityP(A)=∑iP(A∣Bi) P(Bi)P(A) = \sum_i P(A \mid B_i)\,P(B_i)
BayesP(B∣A)=P(A∣B) P(B)P(A)P(B \mid A) = \dfrac{P(A \mid B)\,P(B)}{P(A)}
IndependenceP(A∩B)=P(A) P(B)P(A \cap B) = P(A)\,P(B)

"At least one" is almost always easiest as 1−P(none)1 - P(\text{none}).

Expectation and variance

Linearity of expectation →
FactFormula
Linearity (no independence needed)E[aX+bY]=aE[X]+bE[Y]E[aX + bY] = aE[X] + bE[Y]
VarianceVar⁡(X)=E[X2]−E[X]2\operatorname{Var}(X) = E[X^2] - E[X]^2
SumVar⁡(X+Y)=Var⁡X+Var⁡Y+2Cov⁡(X,Y)\operatorname{Var}(X+Y) = \operatorname{Var}X + \operatorname{Var}Y + 2\operatorname{Cov}(X,Y)
Tower propertyE[X]=E[E[X∣Y]]E[X] = E\big[E[X \mid Y]\big]
Total varianceVar⁡X=E[Var⁡(X∣Y)]+Var⁡(E[X∣Y])\operatorname{Var}X = E[\operatorname{Var}(X\mid Y)] + \operatorname{Var}(E[X \mid Y])
Indicator trickE[∑i1Ai]=∑iP(Ai)E\big[\sum_i \mathbf 1_{A_i}\big] = \sum_i P(A_i)
Tail sum (X≥0X \ge 0 integer)E[X]=∑k≥1P(X≥k)E[X] = \sum_{k \ge 1} P(X \ge k)
DistributionMeanVariance
Bernoulli(p)(p)ppp(1−p)p(1-p)
Binomial(n,p)(n,p)npnpnp(1−p)np(1-p)
Geometric(p)(p), trials to 1st success1/p1/p(1−p)/p2(1-p)/p^2
Poisson(λ)(\lambda)λ\lambdaλ\lambda
Uniform(a,b)(a,b)(a+b)/2(a+b)/2(b−a)2/12(b-a)^2/12
Exponential(λ)(\lambda)1/λ1/\lambda1/λ21/\lambda^2
Normal(μ,σ2)(\mu,\sigma^2)μ\muσ2\sigma^2

Exponential and geometric are memoryless. A fair die: mean 3.53.5, variance 35/1235/12.

ProblemAnswer
Coupon collector, nn typesnHn=n(1+12+⋯+1n)n H_n = n\left(1 + \tfrac12 + \dots + \tfrac1n\right) draws
Fair coin: tosses until HH / until HT66 / 44
Gambler's ruin, fair, start ii, target NNP(reach N)=i/NP(\text{reach } N) = i/N, expected steps i(N−i)i(N-i)
Max of nn iid Uniform(0,1)(0,1)E=nn+1E = \dfrac{n}{n+1}
Stick broken at two uniform points forms a triangle1/41/4
Birthday problem>50%> 50\% shared at 2323 people
Random permutation with no fixed point→1/e≈0.368\to 1/e \approx 0.368
Expected fixed points of a random permutation11

The normal distribution

CLT guide →
WithinProbability
±1σ\pm 1\sigma≈68%\approx 68\%
±2σ\pm 2\sigma≈95%\approx 95\%
±3σ\pm 3\sigma≈99.7%\approx 99.7\%

Two-sided quantiles: 90%→1.64590\% \to 1.645, 95%→1.9695\% \to 1.96, 99%→2.57699\% \to 2.576. Sample mean: standard error σ/n\sigma/\sqrt n.

Mental-math anchors

Mental multiplication →
TrickExample
Difference of squares47×53=502−32=249147 \times 53 = 50^2 - 3^2 = 2491
Square near a base962=(100−4)2=921696^2 = (100-4)^2 = 9216
Ends in 565265^2: 6×7=426 \times 7 = 42, append 2525 →4225\to 4225
Sevenths1/7=0.142857‾1/7 = 0.\overline{142857}
Eighths1/8=0.1251/8 = 0.125, 3/8=0.3753/8 = 0.375, 5/8=0.6255/8 = 0.625
Percent swapx%x\% of yy == y%y\% of xx (e.g. 8%8\% of 25=225 = 2)
Doubling timeabout 72/r72 / r years at r%r\%
FactFormula
Put-call parity (European)C−P=S−Ke−rTC - P = S - K e^{-rT}
ATM call, quick estimateC≈0.4 σSTC \approx 0.4\,\sigma S \sqrt{T}
Call delta (Black-Scholes)ΔC=N(d1)\Delta_C = N(d_1), and ΔP=ΔC−1\Delta_P = \Delta_C - 1
Long call or putgamma >0> 0, vega >0> 0, theta usually <0< 0
Volatility scales with time\sqrt{\text{time}}σdaily≈σannual/252\sigma_{\text{daily}} \approx \sigma_{\text{annual}} / \sqrt{252}

Betting and sizing

Kelly criterion guide →
FactFormula
Expected value of a betEV=p⋅win−(1−p)⋅lossEV = p \cdot \text{win} - (1-p) \cdot \text{loss}
Kelly fraction, odds b:1b{:}1f∗=p−1−pbf^* = p - \dfrac{1-p}{b}
Even-money Kellyf∗=2p−1f^* = 2p - 1
Penalty-scored multiple choice (+1/0/−1)answer only if P(right)>1/2P(\text{right}) > 1/2
FactFormula
Choose(nk)=n!k! (n−k)!\binom{n}{k} = \dfrac{n!}{k!\,(n-k)!}
Stars and bars, nn identical into kk boxes(n+k−1k−1)\binom{n+k-1}{k-1}
Arrangements around a circle(n−1)!(n-1)!
Geometric series, ∣r∣<1\lvert r \rvert < 1∑k≥0rk=11−r\sum_{k\ge0} r^k = \dfrac{1}{1-r}
Harmonic numbersHn≈ln⁡n+0.577H_n \approx \ln n + 0.577
Exponentialex=∑k≥0xk/k!e^x = \sum_{k \ge 0} x^k / k!, and (1−1/n)n→1/e(1 - 1/n)^n \to 1/e

Common questions

What formulas should I memorise for a quant interview?
Bayes' rule and the law of total probability, linearity of expectation, the variance identities, the mean and variance of the common distributions, a few classic results (coupon collector, gambler's ruin, waiting times for coin patterns), the normal 68-95-99.7 rule, put-call parity and the Kelly fraction. Know why each is true; interviewers ask.
Can I use a formula sheet in a quant interview?
Almost never. Live rounds and online assessments are closed-book, which is why the results above are worth memorising rather than looking up.
What is the quickest way to estimate an at-the-money option price?
With rates near zero, an at-the-money call or put is worth about 0.4 × σ × S × √T, where σ is annual volatility, S the spot price and T the time in years.

Knowing the formula is step one

Use it under a clock: free timed simulators, and practice questions with worked solutions for every result on this page.

Read next