Study guides / QNT: Quant Interviews / Statistics & distributions
Statistics & distributions
QNT: Quant Interviews · quick review
The core statistics interviewers expect: normal distribution rules of thumb, variance of sums, standard errors and regression basics.
The rules
- Normal rule of thumb: about 68% within 1 standard deviation, 95% within 2 (more precisely 1.96), 99.7% within 3.
- Variance of a sum of independent variables is the sum of variances; standard deviations don't add.
- Standard error of a mean = σ ÷ √n, so quadrupling the sample halves it.
- Binomial: n trials with probability p have mean np and variance np(1 − p).
- Regression: in simple linear regression, R² is the correlation squared. Correlation doesn't prove causation.
Formulas
- Variance of a sum (independent): Var(X + Y) = Var(X) + Var(Y)
- Standard error: σ ÷ √n
- Binomial variance: n × p × (1 − p)
Common traps
- Adding standard deviations instead of variances.
- Forgetting the square root when converting variance to standard deviation.
- Assuming the mean equals the median in a skewed distribution.
How it's tested
Quant research interviews go deeper, but every quant screen checks these basics.
Practise it
The full lesson, worked examples and practice questions are in Tickmark, where every answer choice is explained.
Open the full lesson