← WTW Interview Insights

WTW·Software Engineer·Technical Phone Screen·Intermediate

Intermediate
Jun 2026

Summary

Quant Engineer interview at WTW with a stats-heavy technical screen. One question but it had some depth to it, more derivation-focused than I expected.

Questions Asked (1)

Q1

Given a normally distributed random variable X with mean mu and variance sigma squared, walk through how you standardize it. Derive Z = (X - mu) / sigma, prove Z follows a standard normal distribution, and explain why standardization actually matters in practice.

Technical Trade-offsAlgorithms & Data Structures
Author's notes

I knew the derivation cold so that part was fine.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by defining the standardization transformation Z = (X - μ)/σ and derive its mean and variance using properties of expectation and variance. Then prove normality by using the linear transformation property of normal distributions or moment generating functions, and finally explain practical benefits like comparability, numerical stability, and enabling use of standard normal tables.

Pro tip: Emphasize that standardization is not just a mathematical trick but a practical necessity for comparing features on different scales, and mention that many machine learning algorithms (e.g., SVM, k-means, PCA) assume or perform better with standardized data.

1. Define the transformation

State that Z = (X - μ)/σ, where X ~ N(μ, σ²). Clarify that this is a linear transformation of X.

2. Compute mean and variance of Z

Use E[Z] = (E[X] - μ)/σ = 0 and Var(Z) = Var(X)/σ² = 1, showing Z has mean 0 and variance 1.

3. Prove Z is normally distributed

Invoke the fact that linear transformations of normal variables are normal, or use the moment generating function: M_Z(t) = exp(t²/2), which is the MGF of a standard normal.

4. Explain practical importance

Discuss how standardization enables comparison across different scales, simplifies probability calculations using Z-tables, and is crucial in algorithms like PCA, SVM, and neural networks.

Key Points to Mention

  • Linearity of expectation and variance properties (Var(aX+b) = a²Var(X))
  • The moment generating function (MGF) of a normal distribution and its uniqueness
  • The standard normal distribution's role in hypothesis testing and confidence intervals
  • Numerical stability and convergence benefits in optimization algorithms
  • Feature scaling in machine learning and its impact on distance-based methods
  • The central limit theorem and why standardization is common in practice

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.